PAWS: Perception of Articulation in the Wild at Scale from Egocentric Videos
Not on the map. Why?
Abstract (arXiv)
Articulation perception aims to recover the motion and structure of articulated objects (e.g., drawers and cupboards), and is fundamental to 3D scene understanding in robotics, simulation, and animation. Existing learning-based methods rely heavily on supervised training with high-quality 3D data and manual annotations, limiting scalability and diversity. To address this limitation, we propose PAWS, a method that directly extracts object articulations from hand-object interactions in large-scale in-the-wild egocentric videos. We evaluate our method on the public data sets, including HD-EPIC and Arti4D data sets, achieving significant improvements over baselines. We further demonstrate that the extracted articulations benefit downstream tasks, including fine-tuning 3D articulation prediction models and enabling robot manipulation. See the project website at https://aaltoml.github.io/PAWS/.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 82 of its 98 references that are in the map. First posted on arXiv 26 Mar 2026.
Builds on
82 of 98 references.
References · 82
- 2023Segment Anythingcited by 8,158
- 2023Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detectioncited by 3,274
- 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,777
- 2021Ego4D: Around the World in 3,000 Hours of Egocentric Videocited by 1,530
- 2024Grounded SAM: Assembling Open-World Models for Diverse Visual Taskscited by 914
- 2024Structured 3D Latents for Scalable and Versatile 3D Generationcited by 742
- 2021DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Camerascited by 673
- 2020SAPIEN: A SimulAted Part-based Interactive ENvironmentcited by 584
- 2022OpenScene: 3D Scene Understanding with Open Vocabulariescited by 556
- 2018PartNet: A Large-scale Benchmark for Fine-grained and Hierarchical Part-level 3D Object Understandingcited by 521
- 2023Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectivescited by 477
- 2024MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motioncited by 431
38 more
- 2020GRAB: A Dataset of Whole-Body Human Grasping of Objectscited by 397
- 2023ConceptGraphs: Open-Vocabulary 3D Scene Graphs for Perception and Planningcited by 357
- 2023Reconstructing Hands in 3D with Transformerscited by 298
- 2025MapAnything: Universal Feed-Forward Metric 3D Reconstructioncited by 295
- 2020Understanding Human Hands in Contact at Internet Scalecited by 281
- 2025MoGe-2: Accurate Monocular Geometry with Metric Scale and Sharp Detailscited by 250
- 2023Affordances from Human Videos as a Versatile Representation for Roboticscited by 221
- 2020ContactPose: A Dataset of Grasps with Object Contact and Hand Posecited by 180
- 2022Ditto: Building Digital Twins of Articulated Objects from Interactioncited by 152
- 2022GAPartNet: Cross-Category Domain-Generalizable Object Perception and Manipulation via Generalizable and Actionable Partscited by 135
- 2022Neural 3D Scene Reconstruction with the Manhattan-world Assumptioncited by 123
- 2019Shape2Motion: Joint Analysis of Motion Parts and Attributes from 3D Shapescited by 117
- 2024WiLoR: End-to-end 3D Hand Localization and Reconstruction in-the-wildcited by 117
- 2025EgoVLA: Learning Vision-Language-Action Models from Egocentric Human Videoscited by 116
- 2023PARIS: Part-level Reconstruction and Motion Analysis for Articulated Objectscited by 115
- 2024URDFormer: A Pipeline for Constructing Articulated Simulation Environments from Real-World Imagescited by 106
- 2025HD-EPIC: A Highly-Detailed Egocentric Video Datasetcited by 101
- 2025Being-H0: Vision-Language-Action Pretraining from Large-Scale Human Videoscited by 94
- 2024Articulate-Anything: Automatic Modeling of Articulated Objects via a Vision-Language Foundation Modelcited by 90
- 2022AKB-48: A Real-World Articulated Object Knowledge Basecited by 81
- 2024General Flow as Foundation Affordance for Scalable Robot Learningcited by 80
- 2022Fine-Grained Egocentric Hand-Object Segmentation: Dataset, Model, and Applicationscited by 78
- 2020ScrewNet: Category-Independent Articulation Model Estimation From Depth Images Using Screw Theorycited by 70
- 2024Neural Implicit Representation for Building Digital Twins of Unknown Articulated Objectscited by 67
- 2025ArtGS: Building Interactable Replicas of Complex Articulated Objects via Gaussian Splattingcited by 67
- 2024SINGAPO: Single Image Controlled Generation of Articulated Parts in Objectscited by 66
- 2025HaWoR: World-Space Hand Motion Reconstruction from Egocentric Videoscited by 62
- 2022DeepLSD: Line Segment Detection and Refinement with Deep Image Gradientscited by 58
- 2024RAM: Retrieval-Based Affordance Transfer for Generalizable Zero-Shot Robotic Manipulationcited by 55
- 2024ParaHome: Parameterizing Everyday Home Activities Towards 3D Generative Modeling of Human-Object Interactionscited by 54
- 2023ArtiGrasp: Physically Plausible Synthesis of Bi-Manual Dexterous Grasping and Articulationcited by 53
- 2025Open-Vocabulary Functional 3D Scene Graphs for Real-World Indoor Spacescited by 53
- 2022OPD: Single-view 3D Openable Part Detectioncited by 52
- 2024Robot See Robot Do: Imitating Articulated Object Manipulation with Monocular 4D Reconstructioncited by 51
- 2023EPIC Fields: Marrying 3D Geometry and Video Understandingcited by 49
- 2025VidBot: Learning Generalizable 3D Actions from In-the-Wild 2D Human Videos for Zero-Shot Robotic Manipulationcited by 46
- 2017From Lifestyle Vlogs to Everyday Interactionscited by 42
- 2025PhysX-3D: Physical-Grounded 3D Asset Generationcited by 40
50 most cited of 82 All 82 (JSON)
Built on by
Citing papers per year: 2026: 1.
1 paper.