Showing 1–33 of 33 results for author: Harley, A W

PointSt3R: Point Tracking through 3D Grounded Correspondence

Authors: Rhodri Guerrier, Adam W. Harley, Dima Damen

Abstract: Recent advances in foundational 3D reconstruction models, such as DUSt3R and MASt3R, have shown great potential in 2D and 3D correspondence in static scenes. In this paper, we propose to adapt them for the task of point tracking through 3D grounded correspondence. We first demonstrate that these models are competitive point trackers when focusing on static points, present in current point tracking… ▽ More Recent advances in foundational 3D reconstruction models, such as DUSt3R and MASt3R, have shown great potential in 2D and 3D correspondence in static scenes. In this paper, we propose to adapt them for the task of point tracking through 3D grounded correspondence. We first demonstrate that these models are competitive point trackers when focusing on static points, present in current point tracking benchmarks ($+33.5\%$ on EgoPoints vs. CoTracker2). We propose to combine the reconstruction loss with training for dynamic correspondence along with a visibility head, and fine-tuning MASt3R for point tracking using a relatively small amount of synthetic data. Importantly, we only train and evaluate on pairs of frames where one contains the query point, effectively removing any temporal context. Using a mix of dynamic and static point correspondences, we achieve competitive or superior point tracking results on four datasets (e.g. competitive on TAP-Vid-DAVIS 73.8 $δ_{avg}$ / 85.8\% occlusion acc. for PointSt3R compared to 75.7 / 88.3\% for CoTracker2; and significantly outperform CoTracker3 on EgoPoints 61.3 vs 54.2 and RGB-S 87.0 vs 82.8). We also present results on 3D point tracking along with several ablations on training datasets and percentage of dynamic correspondences. △ Less

Submitted 30 October, 2025; originally announced October 2025.

Comments: http://rhodriguerrier.github.io/PointSt3R

Generative Point Tracking with Flow Matching

Authors: Mattie Tesfaldet, Adam W. Harley, Konstantinos G. Derpanis, Derek Nowrouzezahrai, Christopher Pal

Abstract: Tracking a point through a video can be a challenging task due to uncertainty arising from visual obfuscations, such as appearance changes and occlusions. Although current state-of-the-art discriminative models excel in regressing long-term point trajectory estimates -- even through occlusions -- they are limited to regressing to a mean (or mode) in the presence of uncertainty, and fail to capture… ▽ More Tracking a point through a video can be a challenging task due to uncertainty arising from visual obfuscations, such as appearance changes and occlusions. Although current state-of-the-art discriminative models excel in regressing long-term point trajectory estimates -- even through occlusions -- they are limited to regressing to a mean (or mode) in the presence of uncertainty, and fail to capture multi-modality. To overcome this limitation, we introduce Generative Point Tracker (GenPT), a generative framework for modelling multi-modal trajectories. GenPT is trained with a novel flow matching formulation that combines the iterative refinement of discriminative trackers, a window-dependent prior for cross-window consistency, and a variance schedule tuned specifically for point coordinates. We show how our model's generative capabilities can be leveraged to improve point trajectory estimates by utilizing a best-first search strategy on generated samples during inference, guided by the model's own confidence of its predictions. Empirically, we evaluate GenPT against the current state of the art on the standard PointOdyssey, Dynamic Replica, and TAP-Vid benchmarks. Further, we introduce a TAP-Vid variant with additional occlusions to assess occluded point tracking performance and highlight our model's ability to capture multi-modality. GenPT is capable of capturing the multi-modality in point trajectories, which translates to state-of-the-art tracking accuracy on occluded points, while maintaining competitive tracking accuracy on visible points compared to extant discriminative point trackers. △ Less

Submitted 23 October, 2025; originally announced October 2025.

Comments: Project page: https://mtesfaldet.net/genpt_projpage/

LookOut: Real-World Humanoid Egocentric Navigation

Authors: Boxiao Pan, Adam W. Harley, C. Karen Liu, Leonidas J. Guibas

Abstract: The ability to predict collision-free future trajectories from egocentric observations is crucial in applications such as humanoid robotics, VR / AR, and assistive navigation. In this work, we introduce the challenging problem of predicting a sequence of future 6D head poses from an egocentric video. In particular, we predict both head translations and rotations to learn the active information-gat… ▽ More The ability to predict collision-free future trajectories from egocentric observations is crucial in applications such as humanoid robotics, VR / AR, and assistive navigation. In this work, we introduce the challenging problem of predicting a sequence of future 6D head poses from an egocentric video. In particular, we predict both head translations and rotations to learn the active information-gathering behavior expressed through head-turning events. To solve this task, we propose a framework that reasons over temporally aggregated 3D latent features, which models the geometric and semantic constraints for both the static and dynamic parts of the environment. Motivated by the lack of training data in this space, we further contribute a data collection pipeline using the Project Aria glasses, and present a dataset collected through this approach. Our dataset, dubbed Aria Navigation Dataset (AND), consists of 4 hours of recording of users navigating in real-world scenarios. It includes diverse situations and navigation behaviors, providing a valuable resource for learning real-world egocentric navigation policies. Extensive experiments show that our model learns human-like navigation behaviors such as waiting / slowing down, rerouting, and looking around for traffic while generalizing to unseen environments. Check out our project webpage at https://sites.google.com/stanford.edu/lookout. △ Less

Submitted 20 August, 2025; originally announced August 2025.

AllTracker: Efficient Dense Point Tracking at High Resolution

Authors: Adam W. Harley, Yang You, Xinglong Sun, Yang Zheng, Nikhil Raghuraman, Yunqi Gu, Sheldon Liang, Wen-Hsuan Chu, Achal Dave, Pavel Tokmakov, Suya You, Rares Ambrus, Katerina Fragkiadaki, Leonidas J. Guibas

Abstract: We introduce AllTracker: a model that estimates long-range point tracks by way of estimating the flow field between a query frame and every other frame of a video. Unlike existing point tracking methods, our approach delivers high-resolution and dense (all-pixel) correspondence fields, which can be visualized as flow maps. Unlike existing optical flow methods, our approach corresponds one frame to… ▽ More We introduce AllTracker: a model that estimates long-range point tracks by way of estimating the flow field between a query frame and every other frame of a video. Unlike existing point tracking methods, our approach delivers high-resolution and dense (all-pixel) correspondence fields, which can be visualized as flow maps. Unlike existing optical flow methods, our approach corresponds one frame to hundreds of subsequent frames, rather than just the next frame. We develop a new architecture for this task, blending techniques from existing work in optical flow and point tracking: the model performs iterative inference on low-resolution grids of correspondence estimates, propagating information spatially via 2D convolution layers, and propagating information temporally via pixel-aligned attention layers. The model is fast and parameter-efficient (16 million parameters), and delivers state-of-the-art point tracking accuracy at high resolution (i.e., tracking 768x1024 pixels, on a 40G GPU). A benefit of our design is that we can train jointly on optical flow datasets and point tracking datasets, and we find that doing so is crucial for top performance. We provide an extensive ablation study on our architecture details and training recipe, making it clear which details matter most. Our code and model weights are available at https://alltracker.github.io △ Less

Submitted 1 August, 2025; v1 submitted 8 June, 2025; originally announced June 2025.

Animal Pose Labeling Using General-Purpose Point Trackers

Authors: Zhuoyang Pan, Boxiao Pan, Guandao Yang, Adam W. Harley, Leonidas Guibas

Abstract: Automatically estimating animal poses from videos is important for studying animal behaviors. Existing methods do not perform reliably since they are trained on datasets that are not comprehensive enough to capture all necessary animal behaviors. However, it is very challenging to collect such datasets due to the large variations in animal morphology. In this paper, we propose an animal pose label… ▽ More Automatically estimating animal poses from videos is important for studying animal behaviors. Existing methods do not perform reliably since they are trained on datasets that are not comprehensive enough to capture all necessary animal behaviors. However, it is very challenging to collect such datasets due to the large variations in animal morphology. In this paper, we propose an animal pose labeling pipeline that follows a different strategy, i.e. test time optimization. Given a video, we fine-tune a lightweight appearance embedding inside a pre-trained general-purpose point tracker on a sparse set of annotated frames. These annotations can be obtained from human labelers or off-the-shelf pose detectors. The fine-tuned model is then applied to the rest of the frames for automatic labeling. Our method achieves state-of-the-art performance at a reasonable annotation cost. We believe our pipeline offers a valuable tool for the automatic quantification of animal behavior. Visit our project webpage at https://zhuoyang-pan.github.io/animal-labeling. △ Less

Submitted 4 June, 2025; originally announced June 2025.

TAPIP3D: Tracking Any Point in Persistent 3D Geometry

Authors: Bowei Zhang, Lei Ke, Adam W. Harley, Katerina Fragkiadaki

Abstract: We introduce TAPIP3D, a novel approach for long-term 3D point tracking in monocular RGB and RGB-D videos. TAPIP3D represents videos as camera-stabilized spatio-temporal feature clouds, leveraging depth and camera motion information to lift 2D video features into a 3D world space where camera movement is effectively canceled out. Within this stabilized 3D representation, TAPIP3D iteratively refines… ▽ More We introduce TAPIP3D, a novel approach for long-term 3D point tracking in monocular RGB and RGB-D videos. TAPIP3D represents videos as camera-stabilized spatio-temporal feature clouds, leveraging depth and camera motion information to lift 2D video features into a 3D world space where camera movement is effectively canceled out. Within this stabilized 3D representation, TAPIP3D iteratively refines multi-frame motion estimates, enabling robust point tracking over long time horizons. To handle the irregular structure of 3D point distributions, we propose a 3D Neighborhood-to-Neighborhood (N2N) attention mechanism - a 3D-aware contextualization strategy that builds informative, spatially coherent feature neighborhoods to support precise trajectory estimation. Our 3D-centric formulation significantly improves performance over existing 3D point tracking methods and even surpasses state-of-the-art 2D pixel trackers in accuracy when reliable depth is available. The model supports inference in both camera-centric (unstabilized) and world-centric (stabilized) coordinates, with experiments showing that compensating for camera motion leads to substantial gains in tracking robustness. By replacing the conventional 2D square correlation windows used in prior 2D and 3D trackers with a spatially grounded 3D attention mechanism, TAPIP3D achieves strong and consistent results across multiple 3D point tracking benchmarks. Project Page: https://tapip3d.github.io △ Less

Submitted 14 November, 2025; v1 submitted 20 April, 2025; originally announced April 2025.

Comments: NeurIPS 2025. Long-term feed-forward 3D point tracking in persistent 3D point maps. Code:https://github.com/zbw001/TAPIP3D

EgoPoints: Advancing Point Tracking for Egocentric Videos

Authors: Ahmad Darkhalil, Rhodri Guerrier, Adam W. Harley, Dima Damen

Abstract: We introduce EgoPoints, a benchmark for point tracking in egocentric videos. We annotate 4.7K challenging tracks in egocentric sequences. Compared to the popular TAP-Vid-DAVIS evaluation benchmark, we include 9x more points that go out-of-view and 59x more points that require re-identification (ReID) after returning to view. To measure the performance of models on these challenging points, we intr… ▽ More We introduce EgoPoints, a benchmark for point tracking in egocentric videos. We annotate 4.7K challenging tracks in egocentric sequences. Compared to the popular TAP-Vid-DAVIS evaluation benchmark, we include 9x more points that go out-of-view and 59x more points that require re-identification (ReID) after returning to view. To measure the performance of models on these challenging points, we introduce evaluation metrics that specifically monitor tracking performance on points in-view, out-of-view, and points that require re-identification. We then propose a pipeline to create semi-real sequences, with automatic ground truth. We generate 11K such sequences by combining dynamic Kubric objects with scene points from EPIC Fields. When fine-tuning point tracking methods on these sequences and evaluating on our annotated EgoPoints sequences, we improve CoTracker across all metrics, including the tracking accuracy $δ^\star_{\text{avg}}$ by 2.7 percentage points and accuracy on ReID sequences (ReID$δ_{\text{avg}}$) by 2.4 points. We also improve $δ^\star_{\text{avg}}$ and ReID$δ_{\text{avg}}$ of PIPs++ by 0.3 and 2.8 respectively. △ Less

Submitted 5 December, 2024; originally announced December 2024.

Comments: Accepted at WACV 2025. Paper webpage: https://ahmaddarkhalil.github.io/EgoPoints/

Monocular Dynamic Gaussian Splatting: Fast, Brittle, and Scene Complexity Rules

Authors: Yiqing Liang, Mikhail Okunev, Mikaela Angelina Uy, Runfeng Li, Leonidas Guibas, James Tompkin, Adam W. Harley

Abstract: Gaussian splatting methods are emerging as a popular approach for converting multi-view image data into scene representations that allow view synthesis. In particular, there is interest in enabling view synthesis for dynamic scenes using only monocular input data -- an ill-posed and challenging problem. The fast pace of work in this area has produced multiple simultaneous papers that claim to work… ▽ More Gaussian splatting methods are emerging as a popular approach for converting multi-view image data into scene representations that allow view synthesis. In particular, there is interest in enabling view synthesis for dynamic scenes using only monocular input data -- an ill-posed and challenging problem. The fast pace of work in this area has produced multiple simultaneous papers that claim to work best, which cannot all be true. In this work, we organize, benchmark, and analyze many Gaussian-splatting-based methods, providing apples-to-apples comparisons that prior works have lacked. We use multiple existing datasets and a new instructive synthetic dataset designed to isolate factors that affect reconstruction quality. We systematically categorize Gaussian splatting methods into specific motion representation types and quantify how their differences impact performance. Empirically, we find that their rank order is well-defined in synthetic data, but the complexity of real-world data currently overwhelms the differences. Furthermore, the fast rendering speed of all Gaussian-based methods comes at the cost of brittleness in optimization. We summarize our experiments into a list of findings that can help to further progress in this lively problem setting. △ Less

Submitted 7 June, 2025; v1 submitted 5 December, 2024; originally announced December 2024.

Comments: TMLR 2025. Project Website: https://brownvc.github.io/MonoDyGauBench.github.io/

View-Consistent Hierarchical 3D Segmentation Using Ultrametric Feature Fields

Authors: Haodi He, Colton Stearns, Adam W. Harley, Leonidas J. Guibas

Abstract: Large-scale vision foundation models such as Segment Anything (SAM) demonstrate impressive performance in zero-shot image segmentation at multiple levels of granularity. However, these zero-shot predictions are rarely 3D-consistent. As the camera viewpoint changes in a scene, so do the segmentation predictions, as well as the characterizations of "coarse" or "fine" granularity. In this work, we ad… ▽ More Large-scale vision foundation models such as Segment Anything (SAM) demonstrate impressive performance in zero-shot image segmentation at multiple levels of granularity. However, these zero-shot predictions are rarely 3D-consistent. As the camera viewpoint changes in a scene, so do the segmentation predictions, as well as the characterizations of "coarse" or "fine" granularity. In this work, we address the challenging task of lifting multi-granular and view-inconsistent image segmentations into a hierarchical and 3D-consistent representation. We learn a novel feature field within a Neural Radiance Field (NeRF) representing a 3D scene, whose segmentation structure can be revealed at different scales by simply using different thresholds on feature distance. Our key idea is to learn an ultrametric feature space, which unlike a Euclidean space, exhibits transitivity in distance-based grouping, naturally leading to a hierarchical clustering. Put together, our method takes view-inconsistent multi-granularity 2D segmentations as input and produces a hierarchy of 3D-consistent segmentations as output. We evaluate our method and several baselines on synthetic datasets with multi-view images and multi-granular segmentation, showcasing improved accuracy and viewpoint-consistency. We additionally provide qualitative examples of our model's 3D hierarchical segmentations in real world scenes. The code and dataset are available at https://github.com/hardyho/ultrametric_feature_fields △ Less

Submitted 17 July, 2024; v1 submitted 30 May, 2024; originally announced May 2024.

ODIN: A Single Model for 2D and 3D Segmentation

Authors: Ayush Jain, Pushkal Katara, Nikolaos Gkanatsios, Adam W. Harley, Gabriel Sarch, Kriti Aggarwal, Vishrav Chaudhary, Katerina Fragkiadaki

Abstract: State-of-the-art models on contemporary 3D segmentation benchmarks like ScanNet consume and label dataset-provided 3D point clouds, obtained through post processing of sensed multiview RGB-D images. They are typically trained in-domain, forego large-scale 2D pre-training and outperform alternatives that featurize the posed RGB-D multiview images instead. The gap in performance between methods that… ▽ More State-of-the-art models on contemporary 3D segmentation benchmarks like ScanNet consume and label dataset-provided 3D point clouds, obtained through post processing of sensed multiview RGB-D images. They are typically trained in-domain, forego large-scale 2D pre-training and outperform alternatives that featurize the posed RGB-D multiview images instead. The gap in performance between methods that consume posed images versus post-processed 3D point clouds has fueled the belief that 2D and 3D perception require distinct model architectures. In this paper, we challenge this view and propose ODIN (Omni-Dimensional INstance segmentation), a model that can segment and label both 2D RGB images and 3D point clouds, using a transformer architecture that alternates between 2D within-view and 3D cross-view information fusion. Our model differentiates 2D and 3D feature operations through the positional encodings of the tokens involved, which capture pixel coordinates for 2D patch tokens and 3D coordinates for 3D feature tokens. ODIN achieves state-of-the-art performance on ScanNet200, Matterport3D and AI2THOR 3D instance segmentation benchmarks, and competitive performance on ScanNet, S3DIS and COCO. It outperforms all previous works by a wide margin when the sensed 3D point cloud is used in place of the point cloud sampled from 3D mesh. When used as the 3D perception engine in an instructable embodied agent architecture, it sets a new state-of-the-art on the TEACh action-from-dialogue benchmark. Our code and checkpoints can be found at the project website (https://odin-seg.github.io). △ Less

Submitted 25 June, 2024; v1 submitted 4 January, 2024; originally announced January 2024.

Comments: Camera Ready (CVPR 2024, Highlight)

Refining Pre-Trained Motion Models

Authors: Xinglong Sun, Adam W. Harley, Leonidas J. Guibas

Abstract: Given the difficulty of manually annotating motion in video, the current best motion estimation methods are trained with synthetic data, and therefore struggle somewhat due to a train/test gap. Self-supervised methods hold the promise of training directly on real video, but typically perform worse. These include methods trained with warp error (i.e., color constancy) combined with smoothness terms… ▽ More Given the difficulty of manually annotating motion in video, the current best motion estimation methods are trained with synthetic data, and therefore struggle somewhat due to a train/test gap. Self-supervised methods hold the promise of training directly on real video, but typically perform worse. These include methods trained with warp error (i.e., color constancy) combined with smoothness terms, and methods that encourage cycle-consistency in the estimates (i.e., tracking backwards should yield the opposite trajectory as tracking forwards). In this work, we take on the challenge of improving state-of-the-art supervised models with self-supervised training. We find that when the initialization is supervised weights, most existing self-supervision techniques actually make performance worse instead of better, which suggests that the benefit of seeing the new data is overshadowed by the noise in the training signal. Focusing on obtaining a "clean" training signal from real-world unlabelled video, we propose to separate label-making and training into two distinct stages. In the first stage, we use the pre-trained model to estimate motion in a video, and then select the subset of motion estimates which we can verify with cycle-consistency. This produces a sparse but accurate pseudo-labelling of the video. In the second stage, we fine-tune the model to reproduce these outputs, while also applying augmentations on the input. We complement this boot-strapping method with simple techniques that densify and re-balance the pseudo-labels, ensuring that we do not merely train on "easy" tracks. We show that our method yields reliable gains over fully-supervised methods in real videos, for both short-term (flow-based) and long-range (multi-frame) pixel tracking. △ Less

Submitted 16 February, 2024; v1 submitted 1 January, 2024; originally announced January 2024.

Comments: Accepted at ICRA 2024

PACE: A Large-Scale Dataset with Pose Annotations in Cluttered Environments

Authors: Yang You, Kai Xiong, Zhening Yang, Zhengxiang Huang, Junwei Zhou, Ruoxi Shi, Zhou Fang, Adam W. Harley, Leonidas Guibas, Cewu Lu

Abstract: We introduce PACE (Pose Annotations in Cluttered Environments), a large-scale benchmark designed to advance the development and evaluation of pose estimation methods in cluttered scenarios. PACE provides a large-scale real-world benchmark for both instance-level and category-level settings. The benchmark consists of 55K frames with 258K annotations across 300 videos, covering 238 objects from 43 c… ▽ More We introduce PACE (Pose Annotations in Cluttered Environments), a large-scale benchmark designed to advance the development and evaluation of pose estimation methods in cluttered scenarios. PACE provides a large-scale real-world benchmark for both instance-level and category-level settings. The benchmark consists of 55K frames with 258K annotations across 300 videos, covering 238 objects from 43 categories and featuring a mix of rigid and articulated items in cluttered scenes. To annotate the real-world data efficiently, we develop an innovative annotation system with a calibrated 3-camera setup. Additionally, we offer PACE-Sim, which contains 100K photo-realistic simulated frames with 2.4M annotations across 931 objects. We test state-of-the-art algorithms in PACE along two tracks: pose estimation, and object pose tracking, revealing the benchmark's challenges and research opportunities. Our benchmark code and data is available on https://github.com/qq456cvb/PACE. △ Less

Submitted 19 July, 2024; v1 submitted 22 December, 2023; originally announced December 2023.

Comments: 14 pages; Accepted to ECCV 2024

Zero-Shot Open-Vocabulary Tracking with Large Pre-Trained Models

Authors: Wen-Hsuan Chu, Adam W. Harley, Pavel Tokmakov, Achal Dave, Leonidas Guibas, Katerina Fragkiadaki

Abstract: Object tracking is central to robot perception and scene understanding. Tracking-by-detection has long been a dominant paradigm for object tracking of specific object categories. Recently, large-scale pre-trained models have shown promising advances in detecting and segmenting objects and parts in 2D static images in the wild. This begs the question: can we re-purpose these large-scale pre-trained… ▽ More Object tracking is central to robot perception and scene understanding. Tracking-by-detection has long been a dominant paradigm for object tracking of specific object categories. Recently, large-scale pre-trained models have shown promising advances in detecting and segmenting objects and parts in 2D static images in the wild. This begs the question: can we re-purpose these large-scale pre-trained static image models for open-vocabulary video tracking? In this paper, we re-purpose an open-vocabulary detector, segmenter, and dense optical flow estimator, into a model that tracks and segments objects of any category in 2D videos. Our method predicts object and part tracks with associated language descriptions in monocular videos, rebuilding the pipeline of Tractor with modern large pre-trained models for static image detection and segmentation: we detect open-vocabulary object instances and propagate their boxes from frame to frame using a flow-based motion model, refine the propagated boxes with the box regression module of the visual detector, and prompt an open-world segmenter with the refined box to segment the objects. We decide the termination of an object track based on the objectness score of the propagated boxes, as well as forward-backward optical flow consistency. We re-identify objects across occlusions using deep feature matching. We show that our model achieves strong performance on multiple established video object segmentation and tracking benchmarks, and can produce reasonable tracks in manipulation data. In particular, our model outperforms previous state-of-the-art in UVO and BURST, benchmarks for open-world object tracking and segmentation, despite never being explicitly trained for tracking. We hope that our approach can serve as a simple and extensible framework for future research. △ Less

Submitted 25 January, 2024; v1 submitted 10 October, 2023; originally announced October 2023.

Comments: Project page available at https://wenhsuanchu.github.io/ovtracktor/

Support-Set Context Matters for Bongard Problems

Authors: Nikhil Raghuraman, Adam W. Harley, Leonidas Guibas

Abstract: Current machine learning methods struggle to solve Bongard problems, which are a type of IQ test that requires deriving an abstract "concept" from a set of positive and negative "support" images, and then classifying whether or not a new query image depicts the key concept. On Bongard-HOI, a benchmark for natural-image Bongard problems, most existing methods have reached at best 69% accuracy (wher… ▽ More Current machine learning methods struggle to solve Bongard problems, which are a type of IQ test that requires deriving an abstract "concept" from a set of positive and negative "support" images, and then classifying whether or not a new query image depicts the key concept. On Bongard-HOI, a benchmark for natural-image Bongard problems, most existing methods have reached at best 69% accuracy (where chance is 50%). Low accuracy is often attributed to neural nets' lack of ability to find human-like symbolic rules. In this work, we point out that many existing methods are forfeiting accuracy due to a much simpler problem: they do not adapt image features given information contained in the support set as a whole, and rely instead on information extracted from individual supports. This is a critical issue, because the "key concept" in a typical Bongard problem can often only be distinguished using multiple positives and multiple negatives. We explore simple methods to incorporate this context and show substantial gains over prior works, leading to new state-of-the-art accuracy on Bongard-LOGO (75.3%) and Bongard-HOI (76.4%) compared to methods with equivalent vision backbone architectures and strong performance on the original Bongard problem set (60.8%). △ Less

Submitted 30 November, 2024; v1 submitted 6 September, 2023; originally announced September 2023.

Comments: TMLR October 2024. Code: https://github.com/nraghuraman/bongard-context

PointOdyssey: A Large-Scale Synthetic Dataset for Long-Term Point Tracking

Authors: Yang Zheng, Adam W. Harley, Bokui Shen, Gordon Wetzstein, Leonidas J. Guibas

Abstract: We introduce PointOdyssey, a large-scale synthetic dataset, and data generation framework, for the training and evaluation of long-term fine-grained tracking algorithms. Our goal is to advance the state-of-the-art by placing emphasis on long videos with naturalistic motion. Toward the goal of naturalism, we animate deformable characters using real-world motion capture data, we build 3D scenes to m… ▽ More We introduce PointOdyssey, a large-scale synthetic dataset, and data generation framework, for the training and evaluation of long-term fine-grained tracking algorithms. Our goal is to advance the state-of-the-art by placing emphasis on long videos with naturalistic motion. Toward the goal of naturalism, we animate deformable characters using real-world motion capture data, we build 3D scenes to match the motion capture environments, and we render camera viewpoints using trajectories mined via structure-from-motion on real videos. We create combinatorial diversity by randomizing character appearance, motion profiles, materials, lighting, 3D assets, and atmospheric effects. Our dataset currently includes 104 videos, averaging 2,000 frames long, with orders of magnitude more correspondence annotations than prior work. We show that existing methods can be trained from scratch in our dataset and outperform the published variants. Finally, we introduce modifications to the PIPs point tracking method, greatly widening its temporal receptive field, which improves its performance on PointOdyssey as well as on two real-world benchmarks. Our data and code are publicly available at: https://pointodyssey.com △ Less

Submitted 27 July, 2023; originally announced July 2023.

TIDEE: Tidying Up Novel Rooms using Visuo-Semantic Commonsense Priors

Authors: Gabriel Sarch, Zhaoyuan Fang, Adam W. Harley, Paul Schydlo, Michael J. Tarr, Saurabh Gupta, Katerina Fragkiadaki

Abstract: We introduce TIDEE, an embodied agent that tidies up a disordered scene based on learned commonsense object placement and room arrangement priors. TIDEE explores a home environment, detects objects that are out of their natural place, infers plausible object contexts for them, localizes such contexts in the current scene, and repositions the objects. Commonsense priors are encoded in three modules… ▽ More We introduce TIDEE, an embodied agent that tidies up a disordered scene based on learned commonsense object placement and room arrangement priors. TIDEE explores a home environment, detects objects that are out of their natural place, infers plausible object contexts for them, localizes such contexts in the current scene, and repositions the objects. Commonsense priors are encoded in three modules: i) visuo-semantic detectors that detect out-of-place objects, ii) an associative neural graph memory of objects and spatial relations that proposes plausible semantic receptacles and surfaces for object repositions, and iii) a visual search network that guides the agent's exploration for efficiently localizing the receptacle-of-interest in the current scene to reposition the object. We test TIDEE on tidying up disorganized scenes in the AI2THOR simulation environment. TIDEE carries out the task directly from pixel and raw depth input without ever having observed the same room beforehand, relying only on priors learned from a separate set of training houses. Human evaluations on the resulting room reorganizations show TIDEE outperforms ablative versions of the model that do not use one or more of the commonsense priors. On a related room rearrangement benchmark that allows the agent to view the goal state prior to rearrangement, a simplified version of our model significantly outperforms a top-performing method by a large margin. Code and data are available at the project website: https://tidee-agent.github.io/. △ Less

Submitted 21 July, 2022; originally announced July 2022.

Simple-BEV: What Really Matters for Multi-Sensor BEV Perception?

Authors: Adam W. Harley, Zhaoyuan Fang, Jie Li, Rares Ambrus, Katerina Fragkiadaki

Abstract: Building 3D perception systems for autonomous vehicles that do not rely on high-density LiDAR is a critical research problem because of the expense of LiDAR systems compared to cameras and other sensors. Recent research has developed a variety of camera-only methods, where features are differentiably "lifted" from the multi-camera images onto the 2D ground plane, yielding a "bird's eye view" (BEV)… ▽ More Building 3D perception systems for autonomous vehicles that do not rely on high-density LiDAR is a critical research problem because of the expense of LiDAR systems compared to cameras and other sensors. Recent research has developed a variety of camera-only methods, where features are differentiably "lifted" from the multi-camera images onto the 2D ground plane, yielding a "bird's eye view" (BEV) feature representation of the 3D space around the vehicle. This line of work has produced a variety of novel "lifting" methods, but we observe that other details in the training setups have shifted at the same time, making it unclear what really matters in top-performing methods. We also observe that using cameras alone is not a real-world constraint, considering that additional sensors like radar have been integrated into real vehicles for years already. In this paper, we first of all attempt to elucidate the high-impact factors in the design and training protocol of BEV perception models. We find that batch size and input resolution greatly affect performance, while lifting strategies have a more modest effect -- even a simple parameter-free lifter works well. Second, we demonstrate that radar data can provide a substantial boost to performance, helping to close the gap between camera-only and LiDAR-enabled systems. We analyze the radar usage details that lead to good performance, and invite the community to re-consider this commonly-neglected part of the sensor platform. △ Less

Submitted 29 September, 2022; v1 submitted 16 June, 2022; originally announced June 2022.

Particle Video Revisited: Tracking Through Occlusions Using Point Trajectories

Authors: Adam W. Harley, Zhaoyuan Fang, Katerina Fragkiadaki

Abstract: Tracking pixels in videos is typically studied as an optical flow estimation problem, where every pixel is described with a displacement vector that locates it in the next frame. Even though wider temporal context is freely available, prior efforts to take this into account have yielded only small gains over 2-frame methods. In this paper, we revisit Sand and Teller's "particle video" approach, an… ▽ More Tracking pixels in videos is typically studied as an optical flow estimation problem, where every pixel is described with a displacement vector that locates it in the next frame. Even though wider temporal context is freely available, prior efforts to take this into account have yielded only small gains over 2-frame methods. In this paper, we revisit Sand and Teller's "particle video" approach, and study pixel tracking as a long-range motion estimation problem, where every pixel is described with a trajectory that locates it in multiple future frames. We re-build this classic approach using components that drive the current state-of-the-art in flow and object tracking, such as dense cost maps, iterative optimization, and learned appearance updates. We train our models using long-range amodal point trajectories mined from existing optical flow data that we synthetically augment with multi-frame occlusions. We test our approach in trajectory estimation benchmarks and in keypoint label propagation tasks, and compare favorably against state-of-the-art optical flow and feature tracking methods. △ Less

Submitted 25 July, 2022; v1 submitted 8 April, 2022; originally announced April 2022.

CoCoNets: Continuous Contrastive 3D Scene Representations

Authors: Shamit Lal, Mihir Prabhudesai, Ishita Mediratta, Adam W. Harley, Katerina Fragkiadaki

Abstract: This paper explores self-supervised learning of amodal 3D feature representations from RGB and RGB-D posed images and videos, agnostic to object and scene semantic content, and evaluates the resulting scene representations in the downstream tasks of visual correspondence, object tracking, and object detection. The model infers a latent3D representation of the scene in the form of 3D feature points… ▽ More This paper explores self-supervised learning of amodal 3D feature representations from RGB and RGB-D posed images and videos, agnostic to object and scene semantic content, and evaluates the resulting scene representations in the downstream tasks of visual correspondence, object tracking, and object detection. The model infers a latent3D representation of the scene in the form of 3D feature points, where each continuous world 3D point is mapped to its corresponding feature vector. The model is trained for contrastive view prediction by rendering 3D feature clouds in queried viewpoints and matching against the 3D feature point cloud predicted from the query view. Notably, the representation can be queried for any 3D location, even if it is not visible from the input view. Our model brings together three powerful ideas of recent exciting research work: 3D feature grids as a neural bottleneck for view prediction, implicit functions for handling resolution limitations of 3D grids, and contrastive learning for unsupervised training of feature representations. We show the resulting 3D visual feature representations effectively scale across objects and scenes, imagine information occluded or missing from the input viewpoints, track objects over time, align semantically related objects in 3D, and improve 3D object detection. We outperform many existing state-of-the-art methods for 3D feature learning and view prediction, which are either limited by 3D grid spatial resolution, do not attempt to build amodal 3D representations, or do not handle combinatorial scene variability due to their non-convolutional bottlenecks. △ Less

Submitted 8 April, 2021; originally announced April 2021.

Track, Check, Repeat: An EM Approach to Unsupervised Tracking

Authors: Adam W. Harley, Yiming Zuo, Jing Wen, Ayush Mangal, Shubhankar Potdar, Ritwick Chaudhry, Katerina Fragkiadaki

Abstract: We propose an unsupervised method for detecting and tracking moving objects in 3D, in unlabelled RGB-D videos. The method begins with classic handcrafted techniques for segmenting objects using motion cues: we estimate optical flow and camera motion, and conservatively segment regions that appear to be moving independently of the background. Treating these initial segments as pseudo-labels, we lea… ▽ More We propose an unsupervised method for detecting and tracking moving objects in 3D, in unlabelled RGB-D videos. The method begins with classic handcrafted techniques for segmenting objects using motion cues: we estimate optical flow and camera motion, and conservatively segment regions that appear to be moving independently of the background. Treating these initial segments as pseudo-labels, we learn an ensemble of appearance-based 2D and 3D detectors, under heavy data augmentation. We use this ensemble to detect new instances of the "moving" type, even if they are not moving, and add these as new pseudo-labels. Our method is an expectation-maximization algorithm, where in the expectation step we fire all modules and look for agreement among them, and in the maximization step we re-train the modules to improve this agreement. The constraint of ensemble agreement helps combat contamination of the generated pseudo-labels (during the E step), and data augmentation helps the modules generalize to yet-unlabelled data (during the M step). We compare against existing unsupervised object discovery and tracking methods, using challenging videos from CATER and KITTI, and show strong improvements over the state-of-the-art. △ Less

Submitted 7 April, 2021; originally announced April 2021.

Move to See Better: Self-Improving Embodied Object Detection

Authors: Zhaoyuan Fang, Ayush Jain, Gabriel Sarch, Adam W. Harley, Katerina Fragkiadaki

Abstract: Passive methods for object detection and segmentation treat images of the same scene as individual samples and do not exploit object permanence across multiple views. Generalization to novel or difficult viewpoints thus requires additional training with lots of annotations. In contrast, humans often recognize objects by simply moving around, to get more informative viewpoints. In this paper, we pr… ▽ More Passive methods for object detection and segmentation treat images of the same scene as individual samples and do not exploit object permanence across multiple views. Generalization to novel or difficult viewpoints thus requires additional training with lots of annotations. In contrast, humans often recognize objects by simply moving around, to get more informative viewpoints. In this paper, we propose a method for improving object detection in testing environments, assuming nothing but an embodied agent with a pre-trained 2D object detector. Our agent collects multi-view data, generates 2D and 3D pseudo-labels, and fine-tunes its detector in a self-supervised manner. Experiments on both indoor and outdoor datasets show that (1) our method obtains high-quality 2D and 3D pseudo-labels from multi-view RGB-D data; (2) fine-tuning with these pseudo-labels improves the 2D detector significantly in the test environment; (3) training a 3D detector with our pseudo-labels outperforms a prior self-supervised method by a large margin; (4) given weak supervision, our method can generate better pseudo-labels for novel objects. △ Less

Submitted 29 March, 2021; v1 submitted 30 November, 2020; originally announced December 2020.

Comments: First three authors contributed equally. Project Page: https://ayushjain1144.github.io/SeeingByMoving/

Disentangling 3D Prototypical Networks For Few-Shot Concept Learning

Authors: Mihir Prabhudesai, Shamit Lal, Darshan Patil, Hsiao-Yu Tung, Adam W Harley, Katerina Fragkiadaki

Abstract: We present neural architectures that disentangle RGB-D images into objects' shapes and styles and a map of the background scene, and explore their applications for few-shot 3D object detection and few-shot concept classification. Our networks incorporate architectural biases that reflect the image formation process, 3D geometry of the world scene, and shape-style interplay. They are trained end-to… ▽ More We present neural architectures that disentangle RGB-D images into objects' shapes and styles and a map of the background scene, and explore their applications for few-shot 3D object detection and few-shot concept classification. Our networks incorporate architectural biases that reflect the image formation process, 3D geometry of the world scene, and shape-style interplay. They are trained end-to-end self-supervised by predicting views in static scenes, alongside a small number of 3D object boxes. Objects and scenes are represented in terms of 3D feature grids in the bottleneck of the network. We show that the proposed 3D neural representations are compositional: they can generate novel 3D scene feature maps by mixing object shapes and styles, resizing and adding the resulting object 3D feature maps over background scene feature maps. We show that classifiers for object categories, color, materials, and spatial relationships trained over the disentangled 3D feature sub-spaces generalize better with dramatically fewer examples than the current state-of-the-art, and enable a visual question answering system that uses them as its modules to generalize one-shot to novel objects in the scene. △ Less

Submitted 20 July, 2021; v1 submitted 6 November, 2020; originally announced November 2020.

3D Object Recognition By Corresponding and Quantizing Neural 3D Scene Representations

Authors: Mihir Prabhudesai, Shamit Lal, Hsiao-Yu Fish Tung, Adam W. Harley, Shubhankar Potdar, Katerina Fragkiadaki

Abstract: We propose a system that learns to detect objects and infer their 3D poses in RGB-D images. Many existing systems can identify objects and infer 3D poses, but they heavily rely on human labels and 3D annotations. The challenge here is to achieve this without relying on strong supervision signals. To address this challenge, we propose a model that maps RGB-D images to a set of 3D visual feature map… ▽ More We propose a system that learns to detect objects and infer their 3D poses in RGB-D images. Many existing systems can identify objects and infer 3D poses, but they heavily rely on human labels and 3D annotations. The challenge here is to achieve this without relying on strong supervision signals. To address this challenge, we propose a model that maps RGB-D images to a set of 3D visual feature maps in a differentiable fully-convolutional manner, supervised by predicting views. The 3D feature maps correspond to a featurization of the 3D world scene depicted in the images. The object 3D feature representations are invariant to camera viewpoint changes or zooms, which means feature matching can identify similar objects under different camera viewpoints. We can compare the 3D feature maps of two objects by searching alignment across scales and 3D rotations, and, as a result of the operation, we can estimate pose and scale changes without the need for 3D pose annotations. We cluster object feature maps into a set of 3D prototypes that represent familiar objects in canonical scales and orientations. We then parse images by inferring the prototype identity and 3D pose for each detected object. We compare our method to numerous baselines that do not learn 3D feature visual representations or do not attempt to correspond features across scenes, and outperform them by a large margin in the tasks of object retrieval and object pose estimation. Thanks to the 3D nature of the object-centric feature maps, the visual similarity cues are invariant to 3D pose changes or small scale changes, which gives our method an advantage over 2D and 1D methods. △ Less

Submitted 30 October, 2020; originally announced October 2020.

Tracking Emerges by Looking Around Static Scenes, with Neural 3D Mapping

Authors: Adam W. Harley, Shrinidhi K. Lakshmikanth, Paul Schydlo, Katerina Fragkiadaki

Abstract: We hypothesize that an agent that can look around in static scenes can learn rich visual representations applicable to 3D object tracking in complex dynamic scenes. We are motivated in this pursuit by the fact that the physical world itself is mostly static, and multiview correspondence labels are relatively cheap to collect in static scenes, e.g., by triangulation. We propose to leverage multivie… ▽ More We hypothesize that an agent that can look around in static scenes can learn rich visual representations applicable to 3D object tracking in complex dynamic scenes. We are motivated in this pursuit by the fact that the physical world itself is mostly static, and multiview correspondence labels are relatively cheap to collect in static scenes, e.g., by triangulation. We propose to leverage multiview data of \textit{static points} in arbitrary scenes (static or dynamic), to learn a neural 3D mapping module which produces features that are correspondable across time. The neural 3D mapper consumes RGB-D data as input, and produces a 3D voxel grid of deep features as output. We train the voxel features to be correspondable across viewpoints, using a contrastive loss, and correspondability across time emerges automatically. At test time, given an RGB-D video with approximate camera poses, and given the 3D box of an object to track, we track the target object by generating a map of each timestep and locating the object's features within each map. In contrast to models that represent video streams in 2D or 2.5D, our model's 3D scene representation is disentangled from projection artifacts, is stable under camera motion, and is robust to partial occlusions. We test the proposed architectures in challenging simulated and real data, and show that our unsupervised 3D object trackers outperform prior unsupervised 2D and 2.5D trackers, and approach the accuracy of supervised trackers. This work demonstrates that 3D object trackers can emerge without tracking labels, through multiview self-supervision on static data. △ Less

Submitted 3 August, 2020; originally announced August 2020.

Embodied Language Grounding with 3D Visual Feature Representations

Authors: Mihir Prabhudesai, Hsiao-Yu Fish Tung, Syed Ashar Javed, Maximilian Sieb, Adam W. Harley, Katerina Fragkiadaki

Abstract: We propose associating language utterances to 3D visual abstractions of the scene they describe. The 3D visual abstractions are encoded as 3-dimensional visual feature maps. We infer these 3D visual scene feature maps from RGB images of the scene via view prediction: when the generated 3D scene feature map is neurally projected from a camera viewpoint, it should match the corresponding RGB image.… ▽ More We propose associating language utterances to 3D visual abstractions of the scene they describe. The 3D visual abstractions are encoded as 3-dimensional visual feature maps. We infer these 3D visual scene feature maps from RGB images of the scene via view prediction: when the generated 3D scene feature map is neurally projected from a camera viewpoint, it should match the corresponding RGB image. We present generative models that condition on the dependency tree of an utterance and generate a corresponding visual 3D feature map as well as reason about its plausibility, and detector models that condition on both the dependency tree of an utterance and a related image and localize the object referents in the 3D feature map inferred from the image. Our model outperforms models of language and vision that associate language with 2D CNN activations or 2D images by a large margin in a variety of tasks, such as, classifying plausibility of utterances, detecting referential expressions, and supplying rewards for trajectory optimization of object placement policies from language instructions. We perform numerous ablations and show the improved performance of our detectors is due to its better generalization across camera viewpoints and lack of object interferences in the inferred 3D feature space, and the improved performance of our generators is due to their ability to spatially reason about objects and their configurations in 3D when mapping from language to scenes. △ Less

Submitted 17 June, 2021; v1 submitted 2 October, 2019; originally announced October 2019.

Journal ref: Conference on Computer Vision and Pattern Recognition. 2020, pp. 2220-2229

Learning from Unlabelled Videos Using Contrastive Predictive Neural 3D Mapping

Authors: Adam W. Harley, Shrinidhi K. Lakshmikanth, Fangyu Li, Xian Zhou, Hsiao-Yu Fish Tung, Katerina Fragkiadaki

Abstract: Predictive coding theories suggest that the brain learns by predicting observations at various levels of abstraction. One of the most basic prediction tasks is view prediction: how would a given scene look from an alternative viewpoint? Humans excel at this task. Our ability to imagine and fill in missing information is tightly coupled with perception: we feel as if we see the world in 3 dimension… ▽ More Predictive coding theories suggest that the brain learns by predicting observations at various levels of abstraction. One of the most basic prediction tasks is view prediction: how would a given scene look from an alternative viewpoint? Humans excel at this task. Our ability to imagine and fill in missing information is tightly coupled with perception: we feel as if we see the world in 3 dimensions, while in fact, information from only the front surface of the world hits our retinas. This paper explores the role of view prediction in the development of 3D visual recognition. We propose neural 3D mapping networks, which take as input 2.5D (color and depth) video streams captured by a moving camera, and lift them to stable 3D feature maps of the scene, by disentangling the scene content from the motion of the camera. The model also projects its 3D feature maps to novel viewpoints, to predict and match against target views. We propose contrastive prediction losses to replace the standard color regression loss, and show that this leads to better performance on complex photorealistic data. We show that the proposed model learns visual representations useful for (1) semi-supervised learning of 3D object detectors, and (2) unsupervised learning of 3D moving object detectors, by estimating the motion of the inferred 3D feature maps in videos of dynamic scenes. To the best of our knowledge, this is the first work that empirically shows view prediction to be a scalable self-supervised task beneficial to 3D object detection. △ Less

Submitted 16 May, 2020; v1 submitted 9 June, 2019; originally announced June 2019.

Image Disentanglement and Uncooperative Re-Entanglement for High-Fidelity Image-to-Image Translation

Authors: Adam W. Harley, Shih-En Wei, Jason Saragih, Katerina Fragkiadaki

Abstract: Cross-domain image-to-image translation should satisfy two requirements: (1) preserve the information that is common to both domains, and (2) generate convincing images covering variations that appear in the target domain. This is challenging, especially when there are no example translations available as supervision. Adversarial cycle consistency was recently proposed as a solution, with beautifu… ▽ More Cross-domain image-to-image translation should satisfy two requirements: (1) preserve the information that is common to both domains, and (2) generate convincing images covering variations that appear in the target domain. This is challenging, especially when there are no example translations available as supervision. Adversarial cycle consistency was recently proposed as a solution, with beautiful and creative results, yielding much follow-up work. However, augmented reality applications cannot readily use such techniques to provide users with compelling translations of real scenes, because the translations do not have high-fidelity constraints. In other words, current models are liable to change details that should be preserved: while re-texturing a face, they may alter the face's expression in an unpredictable way. In this paper, we introduce the problem of high-fidelity image-to-image translation, and present a method for solving it. Our main insight is that low-fidelity translations typically escape a cycle-consistency penalty, because the back-translator learns to compensate for the forward-translator's errors. We therefore introduce an optimization technique that prevents the networks from cooperating: simply train each network only when its input data is real. Prior works, in comparison, train each network with a mix of real and generated data. Experimental results show that our method accurately disentangles the factors that separate the domains, and converges to semantics-preserving translations that prior methods miss. △ Less

Submitted 19 October, 2019; v1 submitted 11 January, 2019; originally announced January 2019.

Reward Learning from Narrated Demonstrations

Authors: Hsiao-Yu Fish Tung, Adam W. Harley, Liang-Kang Huang, Katerina Fragkiadaki

Abstract: Humans effortlessly "program" one another by communicating goals and desires in natural language. In contrast, humans program robotic behaviours by indicating desired object locations and poses to be achieved, by providing RGB images of goal configurations, or supplying a demonstration to be imitated. None of these methods generalize across environment variations, and they convey the goal in awkwa… ▽ More Humans effortlessly "program" one another by communicating goals and desires in natural language. In contrast, humans program robotic behaviours by indicating desired object locations and poses to be achieved, by providing RGB images of goal configurations, or supplying a demonstration to be imitated. None of these methods generalize across environment variations, and they convey the goal in awkward technical terms. This work proposes joint learning of natural language grounding and instructable behavioural policies reinforced by perceptual detectors of natural language expressions, grounded to the sensory inputs of the robotic agent. Our supervision is narrated visual demonstrations(NVD), which are visual demonstrations paired with verbal narration (as opposed to being silent). We introduce a dataset of NVD where teachers perform activities while describing them in detail. We map the teachers' descriptions to perceptual reward detectors, and use them to train corresponding behavioural policies in simulation.We empirically show that our instructable agents (i) learn visual reward detectors using a small number of examples by exploiting hard negative mined configurations from demonstration dynamics, (ii) develop pick-and place policies using learned visual reward detectors, (iii) benefit from object-factorized state representations that mimic the syntactic structure of natural language goal expressions, and (iv) can execute behaviours that involve novel objects in novel locations at test time, instructed by natural language. △ Less

Submitted 27 April, 2018; originally announced April 2018.

Comments: The work has been accepted to Conference on Computer Vision and Pattern Recognition (CVPR) 2018

Segmentation-Aware Convolutional Networks Using Local Attention Masks

Authors: Adam W. Harley, Konstantinos G. Derpanis, Iasonas Kokkinos

Abstract: We introduce an approach to integrate segmentation information within a convolutional neural network (CNN). This counter-acts the tendency of CNNs to smooth information across regions and increases their spatial precision. To obtain segmentation information, we set up a CNN to provide an embedding space where region co-membership can be estimated based on Euclidean distance. We use these embedding… ▽ More We introduce an approach to integrate segmentation information within a convolutional neural network (CNN). This counter-acts the tendency of CNNs to smooth information across regions and increases their spatial precision. To obtain segmentation information, we set up a CNN to provide an embedding space where region co-membership can be estimated based on Euclidean distance. We use these embeddings to compute a local attention mask relative to every neuron position. We incorporate such masks in CNNs and replace the convolution operation with a "segmentation-aware" variant that allows a neuron to selectively attend to inputs coming from its own region. We call the resulting network a segmentation-aware CNN because it adapts its filters at each image point according to local segmentation cues. We demonstrate the merit of our method on two widely different dense prediction tasks, that involve classification (semantic segmentation) and regression (optical flow). Our results show that in semantic segmentation we can match the performance of DenseCRFs while being faster and simpler, and in optical flow we obtain clearly sharper responses than networks that do not use local attention masks. In both cases, segmentation-aware convolution yields systematic improvements over strong baselines. Source code for this work is available online at http://cs.cmu.edu/~aharley/segaware. △ Less

Submitted 15 August, 2017; originally announced August 2017.

Adversarial Inverse Graphics Networks: Learning 2D-to-3D Lifting and Image-to-Image Translation from Unpaired Supervision

Authors: Hsiao-Yu Fish Tung, Adam W. Harley, William Seto, Katerina Fragkiadaki

Abstract: Researchers have developed excellent feed-forward models that learn to map images to desired outputs, such as to the images' latent factors, or to other images, using supervised learning. Learning such mappings from unlabelled data, or improving upon supervised models by exploiting unlabelled data, remains elusive. We argue that there are two important parts to learning without annotations: (i) ma… ▽ More Researchers have developed excellent feed-forward models that learn to map images to desired outputs, such as to the images' latent factors, or to other images, using supervised learning. Learning such mappings from unlabelled data, or improving upon supervised models by exploiting unlabelled data, remains elusive. We argue that there are two important parts to learning without annotations: (i) matching the predictions to the input observations, and (ii) matching the predictions to known priors. We propose Adversarial Inverse Graphics networks (AIGNs): weakly supervised neural network models that combine feedback from rendering their predictions, with distribution matching between their predictions and a collection of ground-truth factors. We apply AIGNs to 3D human pose estimation and 3D structure and egomotion estimation, and outperform models supervised by only paired annotations. We further apply AIGNs to facial image transformation using super-resolution and inpainting renderers, while deliberately adding biases in the ground-truth datasets. Our model seamlessly incorporates such biases, rendering input faces towards young, old, feminine, masculine or Tom Cruise-like equivalents (depending on the chosen bias), or adding lip and nose augmentations while inpainting concealed lips and noses. △ Less

Submitted 1 September, 2017; v1 submitted 31 May, 2017; originally announced May 2017.

Journal ref: The IEEE International Conference on Computer Vision (ICCV), 2017, pp. 4354-4362

Back to Basics: Unsupervised Learning of Optical Flow via Brightness Constancy and Motion Smoothness

Authors: Jason J. Yu, Adam W. Harley, Konstantinos G. Derpanis

Abstract: Recently, convolutional networks (convnets) have proven useful for predicting optical flow. Much of this success is predicated on the availability of large datasets that require expensive and involved data acquisition and laborious la- beling. To bypass these challenges, we propose an unsuper- vised approach (i.e., without leveraging groundtruth flow) to train a convnet end-to-end for predicting o… ▽ More Recently, convolutional networks (convnets) have proven useful for predicting optical flow. Much of this success is predicated on the availability of large datasets that require expensive and involved data acquisition and laborious la- beling. To bypass these challenges, we propose an unsuper- vised approach (i.e., without leveraging groundtruth flow) to train a convnet end-to-end for predicting optical flow be- tween two images. We use a loss function that combines a data term that measures photometric constancy over time with a spatial term that models the expected variation of flow across the image. Together these losses form a proxy measure for losses based on the groundtruth flow. Empiri- cally, we show that a strong convnet baseline trained with the proposed unsupervised approach outperforms the same network trained with supervision on the KITTI dataset. △ Less

Submitted 20 August, 2016; originally announced August 2016.

Learning Dense Convolutional Embeddings for Semantic Segmentation

Authors: Adam W. Harley, Konstantinos G. Derpanis, Iasonas Kokkinos

Abstract: This paper proposes a new deep convolutional neural network (DCNN) architecture that learns pixel embeddings, such that pairwise distances between the embeddings can be used to infer whether or not the pixels lie on the same region. That is, for any two pixels on the same object, the embeddings are trained to be similar; for any pair that straddles an object boundary, the embeddings are trained to… ▽ More This paper proposes a new deep convolutional neural network (DCNN) architecture that learns pixel embeddings, such that pairwise distances between the embeddings can be used to infer whether or not the pixels lie on the same region. That is, for any two pixels on the same object, the embeddings are trained to be similar; for any pair that straddles an object boundary, the embeddings are trained to be dissimilar. Experimental results show that when this embedding network is used in conjunction with a DCNN trained on semantic segmentation, there is a systematic improvement in per-pixel classification accuracy. Our contributions are integrated in the popular Caffe deep learning framework, and consist in straightforward modifications to convolution routines. As such, they can be exploited for any task involving convolution layers. △ Less

Submitted 7 January, 2016; v1 submitted 13 November, 2015; originally announced November 2015.

Evaluation of Deep Convolutional Nets for Document Image Classification and Retrieval

Authors: Adam W. Harley, Alex Ufkes, Konstantinos G. Derpanis

Abstract: This paper presents a new state-of-the-art for document image classification and retrieval, using features learned by deep convolutional neural networks (CNNs). In object and scene analysis, deep neural nets are capable of learning a hierarchical chain of abstraction from pixel inputs to concise and descriptive representations. The current work explores this capacity in the realm of document analy… ▽ More This paper presents a new state-of-the-art for document image classification and retrieval, using features learned by deep convolutional neural networks (CNNs). In object and scene analysis, deep neural nets are capable of learning a hierarchical chain of abstraction from pixel inputs to concise and descriptive representations. The current work explores this capacity in the realm of document analysis, and confirms that this representation strategy is superior to a variety of popular hand-crafted alternatives. Experiments also show that (i) features extracted from CNNs are robust to compression, (ii) CNNs trained on non-document images transfer well to document analysis tasks, and (iii) enforcing region-specific feature-learning is unnecessary given sufficient training data. This work also makes available a new labelled subset of the IIT-CDIP collection, containing 400,000 document images across 16 categories, useful for training new CNNs for document analysis. △ Less

Submitted 25 February, 2015; originally announced February 2015.