Showing 1–36 of 36 results for author: Leroy, V

Spin-wave softening across the uniform-to-stripe domain transition in iron garnet film

Authors: Titiksha Srivastava, Victor Leroy, Nessrine Benaziz, Nathaniel Findling, Ludovic Largeau, Jamal Ben Youssef, Jean-Paul Adam, Thibaut Devolder, Joo-Von Kim

Abstract: Spin-wave spectra across transitions between uniform and textured phases can offer deep insight into both symmetry-breaking physics and self-assembled magnonic bands. However, experiments require a material platform that combines low damping, well-defined textures, and spectroscopic access. Here, we study a Bi-doped iron-garnet film with perpendicular magnetic anisotropy (PMA), which undergoes a u… ▽ More Spin-wave spectra across transitions between uniform and textured phases can offer deep insight into both symmetry-breaking physics and self-assembled magnonic bands. However, experiments require a material platform that combines low damping, well-defined textures, and spectroscopic access. Here, we study a Bi-doped iron-garnet film with perpendicular magnetic anisotropy (PMA), which undergoes a uniform-to-stripe-domain transition as a function of in-plane magnetic field. Real-space imaging by magnetic force microscopy reveals field-reorientable stripe domains aligned with the in-plane field, while reciprocal-space measurements using thermal microfocused Brillouin light scattering ($μ$-BLS) reveal the softening of a low-frequency spin-wave branch near the transition and the appearance of additional modes in the stripe-domain state. Calculated dispersion relations identify finite-$k$ softening in the Damon-Eshbach geometry ($k \perp M$), with the corresponding wavelength matching the stripe periodicity at the transition. In addition, a $μ$-BLS spectral model reproduces the measured mode frequencies and relative intensities at selected fixed fields. Micromagnetic simulations capture the field-driven formation of the stripe state and reproduce the experimental thermal $μ$-BLS spectra. Our findings establish BiYIG with PMA as a model low-damping platform for studying spin-wave freezing, stripe-domain modes, and reconfigurable magnonic band structures. △ Less

Submitted 22 July, 2026; originally announced July 2026.

Comments: Main text: 8 pages, 6 figures. Supplementary Material: 4 pages, 4 figures

Magnonic crystalline properties of stripe textures in thin ferromagnetic films

Authors: Joo-Von Kim, Victor Leroy, Titiksha Srivastava

Abstract: Ordered stripe domains in ferromagnetic thin films form a natural one-dimensional crystal for propagating spin waves. Their spatial periodicity can be tuned readily with an applied magnetic field, making these systems an attractive platform for exploring how magnon band structures evolve as a function of the lattice constant $a$, a tuning that is difficult to achieve in physically patterned materi… ▽ More Ordered stripe domains in ferromagnetic thin films form a natural one-dimensional crystal for propagating spin waves. Their spatial periodicity can be tuned readily with an applied magnetic field, making these systems an attractive platform for exploring how magnon band structures evolve as a function of the lattice constant $a$, a tuning that is difficult to achieve in physically patterned materials. In this work we employ micromagnetic simulations to calculate the spin-wave spectra of a model iron-garnet film, focusing on the influence of the external field and of cubic anisotropy. We find that band gaps at the Brillouin-zone center ($k = 0$) and at the zone boundary ($k = \pm π/a$) respond differently to the applied field, appearing over a broad range of frequencies and wave vectors. When a perpendicular magnetic field or cubic anisotropy is present, additional gaps can appear at the middle of the reduced Brillouin zone ($k = \pm π/2a$). This behavior is interpreted as a Peierls-type distortion of the domain-wall lattice, wherein ``up'' and ``down'' domains alternately expand and contract under the influence of the effective perpendicular field. △ Less

Submitted 21 July, 2026; originally announced July 2026.

Comments: 10 pages, 9 figures

Airborne Minnaert-Like Resonance of an Air-Filled Elasto-Bubble

Authors: Fanambinana Delmotte, Valentin Leroy, Jishen Zhang

Abstract: Deep-subwavelength acoustic resonators are key building blocks of acoustic metamaterials, yet achieving bubble-like resonances in air remains challenging because the Minnaert mechanism relies on the inertia of a surrounding liquid. Here we demonstrate that air-filled soft elastomer shells, termed elasto-bubbles, realize an airborne analogue of the Minnaert resonator. Using impedance-tube measureme… ▽ More Deep-subwavelength acoustic resonators are key building blocks of acoustic metamaterials, yet achieving bubble-like resonances in air remains challenging because the Minnaert mechanism relies on the inertia of a surrounding liquid. Here we demonstrate that air-filled soft elastomer shells, termed elasto-bubbles, realize an airborne analogue of the Minnaert resonator. Using impedance-tube measurements together with the theory of layered-bubble scattering, we show that these soft hollow capsules sustain strong monopolar resonances despite being deeply subwavelength. Their resonance frequency, transmission dip, and absorption are quantitatively captured, without adjustable parameters, by a model accounting for shell elasticity and viscoelasticity. Because shell radius and thickness can be tuned independently during fabrication, elasto-bubbles provide a simple and versatile platform for airborne acoustic metamaterials, resonant absorbers, and acoustic filters. △ Less

Submitted 15 April, 2026; v1 submitted 3 April, 2026; originally announced April 2026.

S-MUSt3R: Sliding Multi-view 3D Reconstruction

Authors: Leonid Antsfeld, Boris Chidlovskii, Yohann Cabon, Vincent Leroy, Jerome Revaud

Abstract: The recent paradigm shift in 3D vision led to the rise of foundation models with remarkable capabilities in 3D perception from uncalibrated images. However, extending these models to large-scale RGB stream 3D reconstruction remains challenging due to memory limitations. This work proposes S-MUSt3R, a simple and efficient pipeline that extends the limits of foundation models for monocular 3D recons… ▽ More The recent paradigm shift in 3D vision led to the rise of foundation models with remarkable capabilities in 3D perception from uncalibrated images. However, extending these models to large-scale RGB stream 3D reconstruction remains challenging due to memory limitations. This work proposes S-MUSt3R, a simple and efficient pipeline that extends the limits of foundation models for monocular 3D reconstruction. Our approach addresses the scalability bottleneck of foundation models through a simple strategy of sequence segmentation followed by segment alignment and lightweight loop closure optimization. Without model retraining, we benefit from remarkable 3D reconstruction capacities of MUSt3R model and achieve trajectory and reconstruction performance comparable to traditional methods with more complex architecture. We evaluate S-MUSt3R on TUM, 7-Scenes and proprietary robot navigation datasets and show that S-MUSt3R runs successfully on long RGB sequences and produces accurate and consistent 3D reconstruction. Our results highlight the potential of leveraging the MUSt3R model for scalable monocular 3D scene in real-world settings, with an important advantage of making predictions directly in the metric space. △ Less

Submitted 4 February, 2026; originally announced February 2026.

Comments: 8 pages, 5 figures, 5 tables

Memory-Based Malware Detection under Limited Data Conditions: A Comparative Evaluation of TabPFN and Ensemble Models

Authors: Valentin Leroy, Shuvalaxmi Dass, Sharif Ullah

Abstract: Artificial intelligence and machine learning have significantly advanced malware research by enabling automated threat detection and behavior analysis. However, the availability of exploitable data is limited, due to the absence of large datasets with real-world data. Despite the progress of AI in cybersecurity, malware analysis still suffers from this data scarcity, which limits model generalizat… ▽ More Artificial intelligence and machine learning have significantly advanced malware research by enabling automated threat detection and behavior analysis. However, the availability of exploitable data is limited, due to the absence of large datasets with real-world data. Despite the progress of AI in cybersecurity, malware analysis still suffers from this data scarcity, which limits model generalization. In order to tackle this difficulty, this workinvestigates TabPFN, a learning-free model designed for low-data regimes. We evaluate its performance against established baselines such as Random Forest, LightGBM and XGBoost, across multiple class configurations. Our experimental results indicate that TabPFN surpasses all other models in low-data regimes, with a 2% to 6% improvement observed across multiple performance metrics. However, this increase in performance has an impact on its computation time in a particular case. These findings highlight both the promise and the practical limitations of integrating TabPFN into cybersecurity workflows. △ Less

Submitted 12 January, 2026; originally announced January 2026.

Comments: 6 pages, 1 figure , 6 Tables

HOSt3R: Keypoint-free Hand-Object 3D Reconstruction from RGB images

Authors: Anilkumar Swamy, Vincent Leroy, Philippe Weinzaepfel, Jean-Sébastien Franco, Grégory Rogez

Abstract: Hand-object 3D reconstruction has become increasingly important for applications in human-robot interaction and immersive AR/VR experiences. A common approach for object-agnostic hand-object reconstruction from RGB sequences involves a two-stage pipeline: hand-object 3D tracking followed by multi-view 3D reconstruction. However, existing methods rely on keypoint detection techniques, such as Struc… ▽ More Hand-object 3D reconstruction has become increasingly important for applications in human-robot interaction and immersive AR/VR experiences. A common approach for object-agnostic hand-object reconstruction from RGB sequences involves a two-stage pipeline: hand-object 3D tracking followed by multi-view 3D reconstruction. However, existing methods rely on keypoint detection techniques, such as Structure from Motion (SfM) and hand-keypoint optimization, which struggle with diverse object geometries, weak textures, and mutual hand-object occlusions, limiting scalability and generalization. As a key enabler to generic and seamless, non-intrusive applicability, we propose in this work a robust, keypoint detector-free approach to estimating hand-object 3D transformations from monocular motion video/images. We further integrate this with a multi-view reconstruction pipeline to accurately recover hand-object 3D shape. Our method, named HOSt3R, is unconstrained, does not rely on pre-scanned object templates or camera intrinsics, and reaches state-of-the-art performance for the tasks of object-agnostic hand-object 3D transformation and shape estimation on the SHOWMe benchmark. We also experiment on sequences from the HO3D dataset, demonstrating generalization to unseen object categories. △ Less

Submitted 25 August, 2025; v1 submitted 22 August, 2025; originally announced August 2025.

Comments: 12 pages, 8 figures

HAMSt3R: Human-Aware Multi-view Stereo 3D Reconstruction

Authors: Sara Rojas, Matthieu Armando, Bernard Ghamen, Philippe Weinzaepfel, Vincent Leroy, Gregory Rogez

Abstract: Recovering the 3D geometry of a scene from a sparse set of uncalibrated images is a long-standing problem in computer vision. While recent learning-based approaches such as DUSt3R and MASt3R have demonstrated impressive results by directly predicting dense scene geometry, they are primarily trained on outdoor scenes with static environments and struggle to handle human-centric scenarios. In this w… ▽ More Recovering the 3D geometry of a scene from a sparse set of uncalibrated images is a long-standing problem in computer vision. While recent learning-based approaches such as DUSt3R and MASt3R have demonstrated impressive results by directly predicting dense scene geometry, they are primarily trained on outdoor scenes with static environments and struggle to handle human-centric scenarios. In this work, we introduce HAMSt3R, an extension of MASt3R for joint human and scene 3D reconstruction from sparse, uncalibrated multi-view images. First, we exploit DUNE, a strong image encoder obtained by distilling, among others, the encoders from MASt3R and from a state-of-the-art Human Mesh Recovery (HMR) model, multi-HMR, for a better understanding of scene geometry and human bodies. Our method then incorporates additional network heads to segment people, estimate dense correspondences via DensePose, and predict depth in human-centric environments, enabling a more comprehensive 3D reconstruction. By leveraging the outputs of our different heads, HAMSt3R produces a dense point map enriched with human semantic information in 3D. Unlike existing methods that rely on complex optimization pipelines, our approach is fully feed-forward and efficient, making it suitable for real-world applications. We evaluate our model on EgoHumans and EgoExo4D, two challenging benchmarks con taining diverse human-centric scenarios. Additionally, we validate its generalization to traditional multi-view stereo and multi-view pose regression tasks. Our results demonstrate that our method can reconstruct humans effectively while preserving strong performance in general 3D reconstruction tasks, bridging the gap between human and scene understanding in 3D vision. △ Less

Submitted 22 August, 2025; originally announced August 2025.

Pow3R: Empowering Unconstrained 3D Reconstruction with Camera and Scene Priors

Authors: Wonbong Jang, Philippe Weinzaepfel, Vincent Leroy, Lourdes Agapito, Jerome Revaud

Abstract: We present Pow3r, a novel large 3D vision regression model that is highly versatile in the input modalities it accepts. Unlike previous feed-forward models that lack any mechanism to exploit known camera or scene priors at test time, Pow3r incorporates any combination of auxiliary information such as intrinsics, relative pose, dense or sparse depth, alongside input images, within a single network.… ▽ More We present Pow3r, a novel large 3D vision regression model that is highly versatile in the input modalities it accepts. Unlike previous feed-forward models that lack any mechanism to exploit known camera or scene priors at test time, Pow3r incorporates any combination of auxiliary information such as intrinsics, relative pose, dense or sparse depth, alongside input images, within a single network. Building upon the recent DUSt3R paradigm, a transformer-based architecture that leverages powerful pre-training, our lightweight and versatile conditioning acts as additional guidance for the network to predict more accurate estimates when auxiliary information is available. During training we feed the model with random subsets of modalities at each iteration, which enables the model to operate under different levels of known priors at test time. This in turn opens up new capabilities, such as performing inference in native image resolution, or point-cloud completion. Our experiments on 3D reconstruction, depth completion, multi-view depth prediction, multi-view stereo, and multi-view pose estimation tasks yield state-of-the-art results and confirm the effectiveness of Pow3r at exploiting all available information. The project webpage is https://europe.naverlabs.com/pow3r. △ Less

Submitted 21 March, 2025; originally announced March 2025.

Comments: CVPR 2025

MUSt3R: Multi-view Network for Stereo 3D Reconstruction

Authors: Yohann Cabon, Lucas Stoffl, Leonid Antsfeld, Gabriela Csurka, Boris Chidlovskii, Jerome Revaud, Vincent Leroy

Abstract: DUSt3R introduced a novel paradigm in geometric computer vision by proposing a model that can provide dense and unconstrained Stereo 3D Reconstruction of arbitrary image collections with no prior information about camera calibration nor viewpoint poses. Under the hood, however, DUSt3R processes image pairs, regressing local 3D reconstructions that need to be aligned in a global coordinate system.… ▽ More DUSt3R introduced a novel paradigm in geometric computer vision by proposing a model that can provide dense and unconstrained Stereo 3D Reconstruction of arbitrary image collections with no prior information about camera calibration nor viewpoint poses. Under the hood, however, DUSt3R processes image pairs, regressing local 3D reconstructions that need to be aligned in a global coordinate system. The number of pairs, growing quadratically, is an inherent limitation that becomes especially concerning for robust and fast optimization in the case of large image collections. In this paper, we propose an extension of DUSt3R from pairs to multiple views, that addresses all aforementioned concerns. Indeed, we propose a Multi-view Network for Stereo 3D Reconstruction, or MUSt3R, that modifies the DUSt3R architecture by making it symmetric and extending it to directly predict 3D structure for all views in a common coordinate frame. Second, we entail the model with a multi-layer memory mechanism which allows to reduce the computational complexity and to scale the reconstruction to large collections, inferring thousands of 3D pointmaps at high frame-rates with limited added complexity. The framework is designed to perform 3D reconstruction both offline and online, and hence can be seamlessly applied to SfM and visual SLAM scenarios showing state-of-the-art performance on various 3D downstream tasks, including uncalibrated Visual Odometry, relative camera pose, scale and focal estimation, 3D reconstruction and multi-view depth estimation. △ Less

Submitted 3 March, 2025; originally announced March 2025.

Comments: Accepted at CVPR 2025

MASt3R-SfM: a Fully-Integrated Solution for Unconstrained Structure-from-Motion

Authors: Bardienus Duisterhof, Lojze Zust, Philippe Weinzaepfel, Vincent Leroy, Yohann Cabon, Jerome Revaud

Abstract: Structure-from-Motion (SfM), a task aiming at jointly recovering camera poses and 3D geometry of a scene given a set of images, remains a hard problem with still many open challenges despite decades of significant progress. The traditional solution for SfM consists of a complex pipeline of minimal solvers which tends to propagate errors and fails when images do not sufficiently overlap, have too l… ▽ More Structure-from-Motion (SfM), a task aiming at jointly recovering camera poses and 3D geometry of a scene given a set of images, remains a hard problem with still many open challenges despite decades of significant progress. The traditional solution for SfM consists of a complex pipeline of minimal solvers which tends to propagate errors and fails when images do not sufficiently overlap, have too little motion, etc. Recent methods have attempted to revisit this paradigm, but we empirically show that they fall short of fixing these core issues. In this paper, we propose instead to build upon a recently released foundation model for 3D vision that can robustly produce local 3D reconstructions and accurate matches. We introduce a low-memory approach to accurately align these local reconstructions in a global coordinate system. We further show that such foundation models can serve as efficient image retrievers without any overhead, reducing the overall complexity from quadratic to linear. Overall, our novel SfM pipeline is simple, scalable, fast and truly unconstrained, i.e. it can handle any collection of images, ordered or not. Extensive experiments on multiple benchmarks show that our method provides steady performance across diverse settings, especially outperforming existing methods in small- and medium-scale settings. △ Less

Submitted 27 September, 2024; originally announced September 2024.

Grounding Image Matching in 3D with MASt3R

Authors: Vincent Leroy, Yohann Cabon, Jérôme Revaud

Abstract: Image Matching is a core component of all best-performing algorithms and pipelines in 3D vision. Yet despite matching being fundamentally a 3D problem, intrinsically linked to camera pose and scene geometry, it is typically treated as a 2D problem. This makes sense as the goal of matching is to establish correspondences between 2D pixel fields, but also seems like a potentially hazardous choice. I… ▽ More Image Matching is a core component of all best-performing algorithms and pipelines in 3D vision. Yet despite matching being fundamentally a 3D problem, intrinsically linked to camera pose and scene geometry, it is typically treated as a 2D problem. This makes sense as the goal of matching is to establish correspondences between 2D pixel fields, but also seems like a potentially hazardous choice. In this work, we take a different stance and propose to cast matching as a 3D task with DUSt3R, a recent and powerful 3D reconstruction framework based on Transformers. Based on pointmaps regression, this method displayed impressive robustness in matching views with extreme viewpoint changes, yet with limited accuracy. We aim here to improve the matching capabilities of such an approach while preserving its robustness. We thus propose to augment the DUSt3R network with a new head that outputs dense local features, trained with an additional matching loss. We further address the issue of quadratic complexity of dense matching, which becomes prohibitively slow for downstream applications if not carefully treated. We introduce a fast reciprocal matching scheme that not only accelerates matching by orders of magnitude, but also comes with theoretical guarantees and, lastly, yields improved results. Extensive experiments show that our approach, coined MASt3R, significantly outperforms the state of the art on multiple matching tasks. In particular, it beats the best published methods by 30% (absolute improvement) in VCRE AUC on the extremely challenging Map-free localization dataset. △ Less

Submitted 14 June, 2024; originally announced June 2024.

DUSt3R: Geometric 3D Vision Made Easy

Authors: Shuzhe Wang, Vincent Leroy, Yohann Cabon, Boris Chidlovskii, Jerome Revaud

Abstract: Multi-view stereo reconstruction (MVS) in the wild requires to first estimate the camera parameters e.g. intrinsic and extrinsic parameters. These are usually tedious and cumbersome to obtain, yet they are mandatory to triangulate corresponding pixels in 3D space, which is the core of all best performing MVS algorithms. In this work, we take an opposite stance and introduce DUSt3R, a radically nov… ▽ More Multi-view stereo reconstruction (MVS) in the wild requires to first estimate the camera parameters e.g. intrinsic and extrinsic parameters. These are usually tedious and cumbersome to obtain, yet they are mandatory to triangulate corresponding pixels in 3D space, which is the core of all best performing MVS algorithms. In this work, we take an opposite stance and introduce DUSt3R, a radically novel paradigm for Dense and Unconstrained Stereo 3D Reconstruction of arbitrary image collections, i.e. operating without prior information about camera calibration nor viewpoint poses. We cast the pairwise reconstruction problem as a regression of pointmaps, relaxing the hard constraints of usual projective camera models. We show that this formulation smoothly unifies the monocular and binocular reconstruction cases. In the case where more than two images are provided, we further propose a simple yet effective global alignment strategy that expresses all pairwise pointmaps in a common reference frame. We base our network architecture on standard Transformer encoders and decoders, allowing us to leverage powerful pretrained models. Our formulation directly provides a 3D model of the scene as well as depth information, but interestingly, we can seamlessly recover from it, pixel matches, relative and absolute camera. Exhaustive experiments on all these tasks showcase that the proposed DUSt3R can unify various 3D vision tasks and set new SoTAs on monocular/multi-view depth estimation as well as relative pose estimation. In summary, DUSt3R makes many geometric 3D vision tasks easy. △ Less

Submitted 2 December, 2024; v1 submitted 21 December, 2023; originally announced December 2023.

Comments: fixing the ref for StaticThings3D dataset

Cross-view and Cross-pose Completion for 3D Human Understanding

Authors: Matthieu Armando, Salma Galaaoui, Fabien Baradel, Thomas Lucas, Vincent Leroy, Romain Brégier, Philippe Weinzaepfel, Grégory Rogez

Abstract: Human perception and understanding is a major domain of computer vision which, like many other vision subdomains recently, stands to gain from the use of large models pre-trained on large datasets. We hypothesize that the most common pre-training strategy of relying on general purpose, object-centric image datasets such as ImageNet, is limited by an important domain shift. On the other hand, colle… ▽ More Human perception and understanding is a major domain of computer vision which, like many other vision subdomains recently, stands to gain from the use of large models pre-trained on large datasets. We hypothesize that the most common pre-training strategy of relying on general purpose, object-centric image datasets such as ImageNet, is limited by an important domain shift. On the other hand, collecting domain-specific ground truth such as 2D or 3D labels does not scale well. Therefore, we propose a pre-training approach based on self-supervised learning that works on human-centric data using only images. Our method uses pairs of images of humans: the first is partially masked and the model is trained to reconstruct the masked parts given the visible ones and a second image. It relies on both stereoscopic (cross-view) pairs, and temporal (cross-pose) pairs taken from videos, in order to learn priors about 3D as well as human motion. We pre-train a model for body-centric tasks and one for hand-centric tasks. With a generic transformer architecture, these models outperform existing self-supervised pre-training methods on a wide set of human-centric downstream tasks, and obtain state-of-the-art performance for instance when fine-tuning for model-based and model-free human mesh recovery. △ Less

Submitted 18 April, 2024; v1 submitted 15 November, 2023; originally announced November 2023.

Comments: CVPR 2024

Win-Win: Training High-Resolution Vision Transformers from Two Windows

Authors: Vincent Leroy, Jerome Revaud, Thomas Lucas, Philippe Weinzaepfel

Abstract: Transformers have become the standard in state-of-the-art vision architectures, achieving impressive performance on both image-level and dense pixelwise tasks. However, training vision transformers for high-resolution pixelwise tasks has a prohibitive cost. Typical solutions boil down to hierarchical architectures, fast and approximate attention, or training on low-resolution crops. This latter so… ▽ More Transformers have become the standard in state-of-the-art vision architectures, achieving impressive performance on both image-level and dense pixelwise tasks. However, training vision transformers for high-resolution pixelwise tasks has a prohibitive cost. Typical solutions boil down to hierarchical architectures, fast and approximate attention, or training on low-resolution crops. This latter solution does not constrain architectural choices, but it leads to a clear performance drop when testing at resolutions significantly higher than that used for training, thus requiring ad-hoc and slow post-processing schemes. In this paper, we propose a novel strategy for efficient training and inference of high-resolution vision transformers. The key principle is to mask out most of the high-resolution inputs during training, keeping only N random windows. This allows the model to learn local interactions between tokens inside each window, and global interactions between tokens from different windows. As a result, the model can directly process the high-resolution input at test time without any special trick. We show that this strategy is effective when using relative positional embedding such as rotary embeddings. It is 4 times faster to train than a full-resolution network, and it is straightforward to use at test time compared to existing approaches. We apply this strategy to three dense prediction tasks with high-resolution data. First, we show on the task of semantic segmentation that a simple setting with 2 windows performs best, hence the name of our method: Win-Win. Second, we confirm this result on the task of monocular depth prediction. Third, we further extend it to the binocular task of optical flow, reaching state-of-the-art performance on the Spring benchmark that contains Full-HD images with an order of magnitude faster inference than the best competitor. △ Less

Submitted 22 March, 2024; v1 submitted 1 October, 2023; originally announced October 2023.

Comments: ICLR 2024

SHOWMe: Benchmarking Object-agnostic Hand-Object 3D Reconstruction

Authors: Anilkumar Swamy, Vincent Leroy, Philippe Weinzaepfel, Fabien Baradel, Salma Galaaoui, Romain Bregier, Matthieu Armando, Jean-Sebastien Franco, Gregory Rogez

Abstract: Recent hand-object interaction datasets show limited real object variability and rely on fitting the MANO parametric model to obtain groundtruth hand shapes. To go beyond these limitations and spur further research, we introduce the SHOWMe dataset which consists of 96 videos, annotated with real and detailed hand-object 3D textured meshes. Following recent work, we consider a rigid hand-object sce… ▽ More Recent hand-object interaction datasets show limited real object variability and rely on fitting the MANO parametric model to obtain groundtruth hand shapes. To go beyond these limitations and spur further research, we introduce the SHOWMe dataset which consists of 96 videos, annotated with real and detailed hand-object 3D textured meshes. Following recent work, we consider a rigid hand-object scenario, in which the pose of the hand with respect to the object remains constant during the whole video sequence. This assumption allows us to register sub-millimetre-precise groundtruth 3D scans to the image sequences in SHOWMe. Although simpler, this hypothesis makes sense in terms of applications where the required accuracy and level of detail is important eg., object hand-over in human-robot collaboration, object scanning, or manipulation and contact point analysis. Importantly, the rigidity of the hand-object systems allows to tackle video-based 3D reconstruction of unknown hand-held objects using a 2-stage pipeline consisting of a rigid registration step followed by a multi-view reconstruction (MVR) part. We carefully evaluate a set of non-trivial baselines for these two stages and show that it is possible to achieve promising object-agnostic 3D hand-object reconstructions employing an SfM toolbox or a hand pose estimator to recover the rigid transforms and off-the-shelf MVR algorithms. However, these methods remain sensitive to the initial camera pose estimates which might be imprecise due to lack of textures on the objects or heavy occlusions of the hands, leaving room for improvements in the reconstruction. Code and dataset are available at https://europe.naverlabs.com/research/showme △ Less

Submitted 19 September, 2023; originally announced September 2023.

Comments: Paper and Appendix, Accepted in ACVR workshop at ICCV conference

4DHumanOutfit: a multi-subject 4D dataset of human motion sequences in varying outfits exhibiting large displacements

Authors: Matthieu Armando, Laurence Boissieux, Edmond Boyer, Jean-Sebastien Franco, Martin Humenberger, Christophe Legras, Vincent Leroy, Mathieu Marsot, Julien Pansiot, Sergi Pujades, Rim Rekik, Gregory Rogez, Anilkumar Swamy, Stefanie Wuhrer

Abstract: This work presents 4DHumanOutfit, a new dataset of densely sampled spatio-temporal 4D human motion data of different actors, outfits and motions. The dataset is designed to contain different actors wearing different outfits while performing different motions in each outfit. In this way, the dataset can be seen as a cube of data containing 4D motion sequences along 3 axes with identity, outfit and… ▽ More This work presents 4DHumanOutfit, a new dataset of densely sampled spatio-temporal 4D human motion data of different actors, outfits and motions. The dataset is designed to contain different actors wearing different outfits while performing different motions in each outfit. In this way, the dataset can be seen as a cube of data containing 4D motion sequences along 3 axes with identity, outfit and motion. This rich dataset has numerous potential applications for the processing and creation of digital humans, e.g. augmented reality, avatar creation and virtual try on. 4DHumanOutfit is released for research purposes at https://kinovis.inria.fr/4dhumanoutfit/. In addition to image data and 4D reconstructions, the dataset includes reference solutions for each axis. We present independent baselines along each axis that demonstrate the value of these reference solutions for evaluation tasks. △ Less

Submitted 12 June, 2023; originally announced June 2023.

CroCo v2: Improved Cross-view Completion Pre-training for Stereo Matching and Optical Flow

Authors: Philippe Weinzaepfel, Thomas Lucas, Vincent Leroy, Yohann Cabon, Vaibhav Arora, Romain Brégier, Gabriela Csurka, Leonid Antsfeld, Boris Chidlovskii, Jérôme Revaud

Abstract: Despite impressive performance for high-level downstream tasks, self-supervised pre-training methods have not yet fully delivered on dense geometric vision tasks such as stereo matching or optical flow. The application of self-supervised concepts, such as instance discrimination or masked image modeling, to geometric tasks is an active area of research. In this work, we build on the recent cross-v… ▽ More Despite impressive performance for high-level downstream tasks, self-supervised pre-training methods have not yet fully delivered on dense geometric vision tasks such as stereo matching or optical flow. The application of self-supervised concepts, such as instance discrimination or masked image modeling, to geometric tasks is an active area of research. In this work, we build on the recent cross-view completion framework, a variation of masked image modeling that leverages a second view from the same scene which makes it well suited for binocular downstream tasks. The applicability of this concept has so far been limited in at least two ways: (a) by the difficulty of collecting real-world image pairs -- in practice only synthetic data have been used -- and (b) by the lack of generalization of vanilla transformers to dense downstream tasks for which relative position is more meaningful than absolute position. We explore three avenues of improvement. First, we introduce a method to collect suitable real-world image pairs at large scale. Second, we experiment with relative positional embeddings and show that they enable vision transformers to perform substantially better. Third, we scale up vision transformer based cross-completion architectures, which is made possible by the use of large amounts of data. With these improvements, we show for the first time that state-of-the-art results on stereo matching and optical flow can be reached without using any classical task-specific techniques like correlation volume, iterative estimation, image warping or multi-scale reasoning, thus paving the way towards universal vision models. △ Less

Submitted 18 August, 2023; v1 submitted 18 November, 2022; originally announced November 2022.

Comments: ICCV 2023

CroCo: Self-Supervised Pre-training for 3D Vision Tasks by Cross-View Completion

Authors: Philippe Weinzaepfel, Vincent Leroy, Thomas Lucas, Romain Brégier, Yohann Cabon, Vaibhav Arora, Leonid Antsfeld, Boris Chidlovskii, Gabriela Csurka, Jérôme Revaud

Abstract: Masked Image Modeling (MIM) has recently been established as a potent pre-training paradigm. A pretext task is constructed by masking patches in an input image, and this masked content is then predicted by a neural network using visible patches as sole input. This pre-training leads to state-of-the-art performance when finetuned for high-level semantic tasks, e.g. image classification and object d… ▽ More Masked Image Modeling (MIM) has recently been established as a potent pre-training paradigm. A pretext task is constructed by masking patches in an input image, and this masked content is then predicted by a neural network using visible patches as sole input. This pre-training leads to state-of-the-art performance when finetuned for high-level semantic tasks, e.g. image classification and object detection. In this paper we instead seek to learn representations that transfer well to a wide variety of 3D vision and lower-level geometric downstream tasks, such as depth prediction or optical flow estimation. Inspired by MIM, we propose an unsupervised representation learning task trained from pairs of images showing the same scene from different viewpoints. More precisely, we propose the pretext task of cross-view completion where the first input image is partially masked, and this masked content has to be reconstructed from the visible content and the second image. In single-view MIM, the masked content often cannot be inferred precisely from the visible portion only, so the model learns to act as a prior influenced by high-level semantics. In contrast, this ambiguity can be resolved with cross-view completion from the second unmasked image, on the condition that the model is able to understand the spatial relationship between the two images. Our experiments show that our pretext task leads to significantly improved performance for monocular 3D vision downstream tasks such as depth estimation. In addition, our model can be directly applied to binocular downstream tasks like optical flow or relative camera pose estimation, for which we obtain competitive results without bells and whistles, i.e., using a generic architecture without any task-specific design. △ Less

Submitted 12 January, 2023; v1 submitted 19 October, 2022; originally announced October 2022.

Comments: NeurIPS 2022

MonoNHR: Monocular Neural Human Renderer

Authors: Hongsuk Choi, Gyeongsik Moon, Matthieu Armando, Vincent Leroy, Kyoung Mu Lee, Gregory Rogez

Abstract: Existing neural human rendering methods struggle with a single image input due to the lack of information in invisible areas and the depth ambiguity of pixels in visible areas. In this regard, we propose Monocular Neural Human Renderer (MonoNHR), a novel approach that renders robust free-viewpoint images of an arbitrary human given only a single image. MonoNHR is the first method that (i) renders… ▽ More Existing neural human rendering methods struggle with a single image input due to the lack of information in invisible areas and the depth ambiguity of pixels in visible areas. In this regard, we propose Monocular Neural Human Renderer (MonoNHR), a novel approach that renders robust free-viewpoint images of an arbitrary human given only a single image. MonoNHR is the first method that (i) renders human subjects never seen during training in a monocular setup, and (ii) is trained in a weakly-supervised manner without geometry supervision. First, we propose to disentangle 3D geometry and texture features and to condition the texture inference on the 3D geometry features. Second, we introduce a Mesh Inpainter module that inpaints the occluded parts exploiting human structural priors such as symmetry. Experiments on ZJU-MoCap, AIST, and HUMBI datasets show that our approach significantly outperforms the recent methods adapted to the monocular case. △ Less

Submitted 2 October, 2022; originally announced October 2022.

Comments: Hongsuk Choi and Gyeongsik Moon contributed equally, 15 pages including the reference and supplementary material

Three-dimensional acoustic lensing with a bubbly diamond metamaterial

Authors: Maxime Lanoy, Fabrice Lemoult, Geoffroy Lerosey, Arnaud Tourin, Valentin Leroy, John H. Page

Abstract: A sound wave travelling in water is scattered by a periodic assembly of air bubbles. The local structure matters even in the low frequency regime. If the bubbles are arranged in a face-centered cubic (fcc) lattice, a total band gap opens near the Minnaert resonance frequency. If they are arranged in the diamond structure, which one obtains by simply adding a second bubble to the unit cell, one fin… ▽ More A sound wave travelling in water is scattered by a periodic assembly of air bubbles. The local structure matters even in the low frequency regime. If the bubbles are arranged in a face-centered cubic (fcc) lattice, a total band gap opens near the Minnaert resonance frequency. If they are arranged in the diamond structure, which one obtains by simply adding a second bubble to the unit cell, one finds an additional branch with a negative slope (optical branch). For a single specific frequency, the medium behaves as if its refractive index (relative to water) is exactly $n=-1$. We show that a slab of this material can be used to design a three three-dimensional flat lens. We also report super-resolution focusing in the near field of the slab and illustrate its potential for imaging in three dimensions. △ Less

Submitted 3 February, 2021; v1 submitted 6 January, 2021; originally announced January 2021.

Comments: 6 pages, 6 figures

SMPLy Benchmarking 3D Human Pose Estimation in the Wild

Authors: Vincent Leroy, Philippe Weinzaepfel, Romain Brégier, Hadrien Combaluzier, Grégory Rogez

Abstract: Predicting 3D human pose from images has seen great recent improvements. Novel approaches that can even predict both pose and shape from a single input image have been introduced, often relying on a parametric model of the human body such as SMPL. While qualitative results for such methods are often shown for images captured in-the-wild, a proper benchmark in such conditions is still missing, as i… ▽ More Predicting 3D human pose from images has seen great recent improvements. Novel approaches that can even predict both pose and shape from a single input image have been introduced, often relying on a parametric model of the human body such as SMPL. While qualitative results for such methods are often shown for images captured in-the-wild, a proper benchmark in such conditions is still missing, as it is cumbersome to obtain ground-truth 3D poses elsewhere than in a motion capture room. This paper presents a pipeline to easily produce and validate such a dataset with accurate ground-truth, with which we benchmark recent 3D human pose estimation methods in-the-wild. We make use of the recently introduced Mannequin Challenge dataset which contains in-the-wild videos of people frozen in action like statues and leverage the fact that people are static and the camera moving to accurately fit the SMPL model on the sequences. A total of 24,428 frames with registered body models are then selected from 567 scenes at almost no cost, using only online RGB videos. We benchmark state-of-the-art SMPL-based human pose estimation methods on this dataset. Our results highlight that challenges remain, in particular for difficult poses or for scenes where the persons are partially truncated or occluded. △ Less

Submitted 4 December, 2020; originally announced December 2020.

Comments: 3DV 2020 Oral presentation

DOPE: Distillation Of Part Experts for whole-body 3D pose estimation in the wild

Authors: Philippe Weinzaepfel, Romain Brégier, Hadrien Combaluzier, Vincent Leroy, Grégory Rogez

Abstract: We introduce DOPE, the first method to detect and estimate whole-body 3D human poses, including bodies, hands and faces, in the wild. Achieving this level of details is key for a number of applications that require understanding the interactions of the people with each other or with the environment. The main challenge is the lack of in-the-wild data with labeled whole-body 3D poses. In previous wo… ▽ More We introduce DOPE, the first method to detect and estimate whole-body 3D human poses, including bodies, hands and faces, in the wild. Achieving this level of details is key for a number of applications that require understanding the interactions of the people with each other or with the environment. The main challenge is the lack of in-the-wild data with labeled whole-body 3D poses. In previous work, training data has been annotated or generated for simpler tasks focusing on bodies, hands or faces separately. In this work, we propose to take advantage of these datasets to train independent experts for each part, namely a body, a hand and a face expert, and distill their knowledge into a single deep network designed for whole-body 2D-3D pose detection. In practice, given a training image with partial or no annotation, each part expert detects its subset of keypoints in 2D and 3D and the resulting estimations are combined to obtain whole-body pseudo ground-truth poses. A distillation loss encourages the whole-body predictions to mimic the experts' outputs. Our results show that this approach significantly outperforms the same whole-body model trained without distillation while staying close to the performance of the experts. Importantly, DOPE is computationally less demanding than the ensemble of experts and can achieve real-time performance. Test code and models are available at https://europe.naverlabs.com/research/computer-vision/dope. △ Less

Submitted 21 August, 2020; originally announced August 2020.

Comments: ECCV 2020

Robust Image Retrieval-based Visual Localization using Kapture

Authors: Martin Humenberger, Yohann Cabon, Nicolas Guerin, Julien Morat, Vincent Leroy, Jérôme Revaud, Philippe Rerole, Noé Pion, Cesar de Souza, Gabriela Csurka

Abstract: Visual localization tackles the challenge of estimating the camera pose from images by using correspondence analysis between query images and a map. This task is computation and data intensive which poses challenges on thorough evaluation of methods on various datasets. However, in order to further advance in the field, we claim that robust visual localization algorithms should be evaluated on mul… ▽ More Visual localization tackles the challenge of estimating the camera pose from images by using correspondence analysis between query images and a map. This task is computation and data intensive which poses challenges on thorough evaluation of methods on various datasets. However, in order to further advance in the field, we claim that robust visual localization algorithms should be evaluated on multiple datasets covering a broad domain variety. To facilitate this, we introduce kapture, a new, flexible, unified data format and toolbox for visual localization and structure-from-motion (SFM). It enables easy usage of different datasets as well as efficient and reusable data processing. To demonstrate this, we present a versatile pipeline for visual localization that facilitates the use of different local and global features, 3D data (e.g. depth maps), non-vision sensor data (e.g. IMU, GPS, WiFi), and various processing algorithms. Using multiple configurations of the pipeline, we show the great versatility of kapture in our experiments. Furthermore, we evaluate our methods on eight public datasets where they rank top on all and first on many of them. To foster future research, we release code, models, and all datasets used in this paper in the kapture format open source under a permissive BSD license. github.com/naver/kapture, github.com/naver/kapture-localization △ Less

Submitted 7 January, 2022; v1 submitted 27 July, 2020; originally announced July 2020.

Mutation++: MUlticomponent Thermodynamic And Transport properties for IONized gases in C++

Authors: James B. Scoggins, Vincent Leroy, Georgios Bellas-Chatzigeorgis, Bruno Dias, Thierry E. Magin

Abstract: The Mutation++ library provides accurate and efficient computation of physicochemical properties associated with partially ionized gases in various degrees of thermal nonequilibrium. With v1.0.0, users can compute thermodynamic and transport properties, multiphase linearly-constrained equilibria, chemical production rates, energy transfer rates, and gas-surface interactions. The framework is based… ▽ More The Mutation++ library provides accurate and efficient computation of physicochemical properties associated with partially ionized gases in various degrees of thermal nonequilibrium. With v1.0.0, users can compute thermodynamic and transport properties, multiphase linearly-constrained equilibria, chemical production rates, energy transfer rates, and gas-surface interactions. The framework is based on an object-oriented design in C++, allowing users to plug-and-play various models, algorithms, and data as necessary. Mutation++ is available open-source under the GNU Lesser General Public License v3.0. △ Less

Submitted 5 February, 2020; originally announced February 2020.

Comments: Mutation++ is available open-source at https://github.com/mutationpp/Mutationpp

Ultrasound transmission through monodisperse 2D microfoams

Authors: Lorène Champougny, Juliette Pierre, Antoine Devulder, Valentin Leroy, Marie-Caroline Jullien

Abstract: While the acoustic properties of solid foams have been abundantly characterized, sound propagation in liquid foams remains poorly understood. Recent studies have investigated the transmission of ultrasound through three-dimensional polydisperse liquid foams (Pierre et al., 2013, 2014, 2017). However, further progress requires to characterize the acoustic response of better controlled foam structur… ▽ More While the acoustic properties of solid foams have been abundantly characterized, sound propagation in liquid foams remains poorly understood. Recent studies have investigated the transmission of ultrasound through three-dimensional polydisperse liquid foams (Pierre et al., 2013, 2014, 2017). However, further progress requires to characterize the acoustic response of better controlled foam structures. In this work, we study experimentally the transmission of ultrasounds through a single layer of monodisperse bubbles generated by microfluidics techniques. In such a material, we show that the sound velocity is only sensitive to the gas phase. Nevertheless, the structure of the liquid network has to be taken into account through a transfer parameter analogous to the one in a layer of porous material. Finally, we observe that the attenuation cannot be explained by thermal dissipation alone, but is compatible with viscous dissipation in the gas pores of the monolayer. △ Less

Submitted 18 January, 2019; originally announced January 2019.

Journal ref: European Physical Journal E (2019) 42: 6

Using Longitudinal Targeted Maximum Likelihood Estimation in Complex Settings with Dynamic Interventions

Authors: Michael Schomaker, Miguel Angel Luque-Fernandez, Valeriane Leroy, Mary-Ann Davies

Abstract: Longitudinal targeted maximum likelihood estimation (LTMLE) has very rarely been used to estimate dynamic treatment effects in the context of time-dependent confounding affected by prior treatment when faced with long follow-up times, multiple time-varying confounders, and complex associational relationships simultaneously. Reasons for this include the potential computational burden, technical cha… ▽ More Longitudinal targeted maximum likelihood estimation (LTMLE) has very rarely been used to estimate dynamic treatment effects in the context of time-dependent confounding affected by prior treatment when faced with long follow-up times, multiple time-varying confounders, and complex associational relationships simultaneously. Reasons for this include the potential computational burden, technical challenges, restricted modeling options for long follow-up times, and limited practical guidance in the literature. However, LTMLE has desirable asymptotic properties, i.e. it is doubly robust, and can yield valid inference when used in conjunction with machine learning. We use a topical and sophisticated question from HIV treatment research to show that LTMLE can be used successfully in complex realistic settings and compare results to competing estimators. Our example illustrates the following practical challenges common to many epidemiological studies 1) long follow-up time (30 months), 2) gradually declining sample size 3) limited support for some intervention rules of interest 4) a high-dimensional set of potential adjustment variables, increasing both the need and the challenge of integrating appropriate machine learning methods 5) consideration of collider bias. Our analyses, as well as simulations, shed new light on the application of LTMLE in complex and realistic settings: we show that (i) LTMLE can yield stable and good estimates, even when confronted with small samples and limited modeling options; (ii) machine learning utilized with a small set of simple learners (if more complex ones can't be fitted) can outperform a single, complex model, which is tailored to incorporate prior clinical knowledge; (iii) performance can vary considerably depending on interventions and their support in the data, and therefore critical quality checks should accompany every LTMLE analysis. △ Less

Submitted 4 March, 2021; v1 submitted 14 February, 2018; originally announced February 2018.

Acoustic double negativity induced by position correlations within a disordered set of monopolar resonators

Authors: Maxime Lanoy, John H. Page, Geoffroy Lerosey, Fabrice Lemoult, Arnaud Tourin, Valentin Leroy

Abstract: Using a Multiple Scattering Theory algorithm, we investigate numerically the transmission of ultrasonic waves through a disordered locally resonant metamaterial containing only monopolar resonators. By comparing the cases of a perfectly random medium with its pair correlated counterpart, we show that the introduction of short range correlation can substantially impact the effective parameters of t… ▽ More Using a Multiple Scattering Theory algorithm, we investigate numerically the transmission of ultrasonic waves through a disordered locally resonant metamaterial containing only monopolar resonators. By comparing the cases of a perfectly random medium with its pair correlated counterpart, we show that the introduction of short range correlation can substantially impact the effective parameters of the sample. We report, notably, the opening of an acoustic transparency window in the region of the hybridization band gap. Interestingly, the transparency window is found to be associated with negative values of both effective compressibility and density. Despite this feature being unexpected for a disordered medium of monopolar resonators, we show that it can be fully described analytically and that it gives rise to negative refraction of waves. △ Less

Submitted 11 October, 2017; v1 submitted 9 October, 2017; originally announced October 2017.

Journal ref: Phys. Rev. B 96, 220201 (2017)

An FFT approach to the analysis of dynamic properties of gas/liquid interfaces

Authors: Sandrine Mariot, Valentin Leroy, Juliette Pierre, Florence Elias, Eloise Bouthemy, Dominique Langevin, Wiebke Drenckhan

Abstract: The characterisation of the dynamic properties of viscoelastic monolayers of surfactants at the gasliquid interface is very important in the analysis and prediction of foam stability. With most of the relevant dynamic processes being rapid (thermal fluctuation, film coalescence etc.) it is important to probe interfacial dynamics at high deformation rates. Today, only few techniques allow this, one… ▽ More The characterisation of the dynamic properties of viscoelastic monolayers of surfactants at the gasliquid interface is very important in the analysis and prediction of foam stability. With most of the relevant dynamic processes being rapid (thermal fluctuation, film coalescence etc.) it is important to probe interfacial dynamics at high deformation rates. Today, only few techniques allow this, one of them being the characterisation of the propagation of electrocapillary waves on the liquid surface. Traditionally, this technique has been applied in a continuous mode (i.e. at constant frequency) in order to ensure reliable accuracy. Here we explore the possibility to analyse the propagation of an excited pulse in order to access the interfacial properties in one single Fourier treatment over a wide range of frequencies. The main advantage of this approach is that the measurement times and the required liquid volumes can be reduced significantly. This occurs at the cost of precision in the measurement, due partly to the presence of a pronounced resonance of the liquid surface. The pulsed approach may therefore be used to prescan the surface response before a more in-depth scan at constant frequency; or to follow the changes of the interfacial properties during surfactant adsorption. △ Less

Submitted 21 October, 2016; originally announced October 2016.

Comments: 17 pages,9 figures

Ternary Neural Networks for Resource-Efficient AI Applications

Authors: Hande Alemdar, Vincent Leroy, Adrien Prost-Boucle, Frédéric Pétrot

Abstract: The computation and storage requirements for Deep Neural Networks (DNNs) are usually high. This issue limits their deployability on ubiquitous computing devices such as smart phones, wearables and autonomous drones. In this paper, we propose ternary neural networks (TNNs) in order to make deep learning more resource-efficient. We train these TNNs using a teacher-student approach based on a novel,… ▽ More The computation and storage requirements for Deep Neural Networks (DNNs) are usually high. This issue limits their deployability on ubiquitous computing devices such as smart phones, wearables and autonomous drones. In this paper, we propose ternary neural networks (TNNs) in order to make deep learning more resource-efficient. We train these TNNs using a teacher-student approach based on a novel, layer-wise greedy methodology. Thanks to our two-stage training procedure, the teacher network is still able to use state-of-the-art methods such as dropout and batch normalization to increase accuracy and reduce training time. Using only ternary weights and activations, the student ternary network learns to mimic the behavior of its teacher network without using any multiplication. Unlike its -1,1 binary counterparts, a ternary neural network inherently prunes the smaller weights by setting them to zero during training. This makes them sparser and thus more energy-efficient. We design a purpose-built hardware architecture for TNNs and implement it on FPGA and ASIC. We evaluate TNNs on several benchmark datasets and demonstrate up to 3.1x better energy efficiency with respect to the state of the art while also improving accuracy. △ Less

Submitted 26 February, 2017; v1 submitted 1 September, 2016; originally announced September 2016.

Testing Interestingness Measures in Practice: A Large-Scale Analysis of Buying Patterns

Authors: Martin Kirchgessner, Vincent Leroy, Sihem Amer-Yahia, Shashwat Mishra

Abstract: Understanding customer buying patterns is of great interest to the retail industry and has shown to benefit a wide variety of goals ranging from managing stocks to implementing loyalty programs. Association rule mining is a common technique for extracting correlations such as "people in the South of France buy rosé wine" or "customers who buy paté also buy salted butter and sour bread." Unfortunat… ▽ More Understanding customer buying patterns is of great interest to the retail industry and has shown to benefit a wide variety of goals ranging from managing stocks to implementing loyalty programs. Association rule mining is a common technique for extracting correlations such as "people in the South of France buy rosé wine" or "customers who buy paté also buy salted butter and sour bread." Unfortunately, sifting through a high number of buying patterns is not useful in practice, because of the predominance of popular products in the top rules. As a result, a number of "interestingness" measures (over 30) have been proposed to rank rules. However, there is no agreement on which measures are more appropriate for retail data. Moreover, since pattern mining algorithms output thousands of association rules for each product, the ability for an analyst to rely on ranking measures to identify the most interesting ones is crucial. In this paper, we develop CAPA (Comparative Analysis of PAtterns), a framework that provides analysts with the ability to compare the outcome of interestingness measures applied to buying patterns in the retail industry. We report on how we used CAPA to compare 34 measures applied to over 1,800 stores of Intermarché, one of the largest food retailers in France. △ Less

Submitted 15 March, 2016; originally announced March 2016.

Time reversal sub-wavelength focusing in bubbly media

Authors: Maxime Lanoy, Romain Pierrat, Fabrice Lemoult, Mathias Fink, Valentin Leroy, Arnaud Tourin

Abstract: Thanks to a Multiple Scattering Theory algorithm, we present a way to focus energy at the deep subwavelength scale, from the far-field, inside a cubic disordered bubble cloud by using broadband Time Reversal (TR). We show that the analytical calculation of an effective wavenumber performing the Independant Scattering Approximation (ISA) matches the numerical results for the focal extension. Subwav… ▽ More Thanks to a Multiple Scattering Theory algorithm, we present a way to focus energy at the deep subwavelength scale, from the far-field, inside a cubic disordered bubble cloud by using broadband Time Reversal (TR). We show that the analytical calculation of an effective wavenumber performing the Independant Scattering Approximation (ISA) matches the numerical results for the focal extension. Subwavelength focusings of lambda/100 are reported for simulations with perfect bubbles (no loss). A more realistic case, with viscous and thermal losses, allows us to obtain a $λ/14$ focal spot, with a low volume fraction of scatterers (phi = 0.01). Bubbly materials could open new perspective for acoustic actuation in the microfluidic context. △ Less

Submitted 23 October, 2015; originally announced October 2015.

Comments: 7 pages, 6 figures

Journal ref: Phys. Rev. B 91, 224202 (2015)

Manipulating bubbles with secondary Bjerknes forces

Authors: Maxime Lanoy, Caroline Derec, Arnaud Tourin, Valentin Leroy

Abstract: Gas bubbles in a sound field are submitted to a radiative force, known as the secondary Bjerknes force. We propose an original experimental setup that allows us to investigate in details this force between two bubbles, as a function of the sonication frequency, as well as the bubbles radii and distance. We report the observation of both attractive and, more interestingly, repulsive Bjerknes force,… ▽ More Gas bubbles in a sound field are submitted to a radiative force, known as the secondary Bjerknes force. We propose an original experimental setup that allows us to investigate in details this force between two bubbles, as a function of the sonication frequency, as well as the bubbles radii and distance. We report the observation of both attractive and, more interestingly, repulsive Bjerknes force, when the two bubbles are driven in antiphase. Our experiments show the importance of taking multiple scattering into account, which leads to a strong acoustic coupling of the bubbles when their radii are similar. Our setup demonstrates the accuracy of secondary Bjerknes forces for attracting or repealing a bubble, and could lead to new acoustic tools for non contact manipulation in microfluidic devices. △ Less

Submitted 23 October, 2015; originally announced October 2015.

Comments: 4 pages, 6 figures

A technique for measuring velocity and attenuation of ultrasound in liquid foams

Authors: Juliette Pierre, F. Elias, Valentin Leroy

Abstract: We describe an experimental setup specifically designed for measuring the ultrasonic transmission through liquid foams, over a broad range of frequencies (60-600 kHz). The question of determining the ultrasonic properties of the foam (density, phase velocity and attenuation) from the transmission measurements is addressed. An inversion method is proposed, tested on synthetic data, and applied to a… ▽ More We describe an experimental setup specifically designed for measuring the ultrasonic transmission through liquid foams, over a broad range of frequencies (60-600 kHz). The question of determining the ultrasonic properties of the foam (density, phase velocity and attenuation) from the transmission measurements is addressed. An inversion method is proposed, tested on synthetic data, and applied to a liquid foam at different times during the coarsening. The ultrasonic velocity and attenuation are found to be very sensitive to the foam bubble sizes, suggesting that a spectroscopy technique could be developed for liquid foams. △ Less

Submitted 19 October, 2012; originally announced October 2012.

Journal ref: Ultrasonics 53 (2013) pp. 622-629

Enhanced and reduced transmission of acoustic waves with bubble meta-screens

Authors: Alice Bretagne, Arnaud Tourin, V. Leroy

Abstract: We present a class of sonic meta-screens for manipulating air-borne acoustic waves at ultrasonic or audible frequencies. Our screens consist of periodic arrangements of air bubbles in water or possibly embedded in a soft elastic matrix. They can be used for soundproofing, but also for exalting transmission at an air/water interface or even to achieve enhanced absorption. We present a class of sonic meta-screens for manipulating air-borne acoustic waves at ultrasonic or audible frequencies. Our screens consist of periodic arrangements of air bubbles in water or possibly embedded in a soft elastic matrix. They can be used for soundproofing, but also for exalting transmission at an air/water interface or even to achieve enhanced absorption. △ Less

Submitted 5 November, 2011; originally announced November 2011.

Influence of positional correlations on the propagation of waves in a complex medium with polydisperse resonant scatterers

Authors: V. Leroy, A. L. Strybulevych, J. H. Page, M. G. Scanlon

Abstract: We present experimental results on a model system for studying wave propagation in a complex medium exhibiting low frequency resonances. These experiments enable us to investigate a fundamental question that is relevant for many materials, such as metamaterials, where low-frequency scattering resonances strongly influence the effective medium properties. This question concerns the effect of correl… ▽ More We present experimental results on a model system for studying wave propagation in a complex medium exhibiting low frequency resonances. These experiments enable us to investigate a fundamental question that is relevant for many materials, such as metamaterials, where low-frequency scattering resonances strongly influence the effective medium properties. This question concerns the effect of correlations in the positions of the scatterers on the coupling between their resonances, and hence on wave transport through the medium. To examine this question experimentally, we measure the effective medium wave number of acoustic waves in a sample made of bubbles embedded in an elastic matrix over a frequency range that includes the resonance frequency of the bubbles. The effective medium is highly dispersive, showing peaks in the attenuation and the phase velocity as functions of the frequency, which cannot be accurately described using the Independent Scattering Approximation (ISA). This discrepancy may be explained by the effects of the positional correlations of the scatterers, which we show to be dependent on the size of the scatterers. We propose a self-consistent approach for taking this "polydisperse correlation" into account and show that our model better describes the experimental results than the ISA. △ Less

Submitted 26 April, 2011; v1 submitted 29 October, 2010; originally announced October 2010.

Journal ref: Physical Review. e, statistical physics, plasmas, fluids, and related interdisciplinary topics (2011) 046605

Thermal degradation of ligno-cellulosic fuels: DSC and TGA studies

Authors: Valérie Leroy, Dominique Cancellieri, Eric Leoni

Abstract: The scope of this work was to show the utility of thermal analysis and calorimetric experiments to study the thermal oxidative degradation of Mediterranean scrubs. We investigated the thermal degradation of four species; DSC and TGA were used under air sweeping to record oxidative reactions in dynamic conditions. Heat released and mass loss are important data to be measured for wildland fires mo… ▽ More The scope of this work was to show the utility of thermal analysis and calorimetric experiments to study the thermal oxidative degradation of Mediterranean scrubs. We investigated the thermal degradation of four species; DSC and TGA were used under air sweeping to record oxidative reactions in dynamic conditions. Heat released and mass loss are important data to be measured for wildland fires modelling purpose and fire hazard studies on ligno-cellulosic fuels. Around 638 and 778 K, two dominating and overlapped exothermic peaks were recorded in DSC and individualized using a experimental and numerical separation. This stage allowed obtaining the enthalpy variation of each exothermic phenomenon. As an application, we propose to classify the fuels according to the heat released and the rate constant of each reaction. TGA experiments showed under air two successive mass loss around 638 and 778 K. Both techniques are useful in order to measure ignitability, combustibility and sustainability of forest fuels. △ Less

Submitted 17 November, 2008; originally announced November 2008.

Journal ref: Thermochimica Acta 451 (2006) 131-138