AnyLift: Scaling Motion Reconstruction from Internet Videos via 2D Diffusion

• URL: http://arxiv.org/abs/2604.17818v1
• Date: Mon, 20 Apr 2026 05:15:38 GMT
• Title: AnyLift: Scaling Motion Reconstruction from Internet Videos via 2D Diffusion
• Abstract summary: We introduce a two-stage framework leveraging 2D diffusion that reconstructs 3D human motion and HOI from Internet videos.<n>In the first stage, we synthesize multi-view 2D motion data for each domain, leveraging 2D keypoints extracted from Internet videos.<n>In the second stage, a camera-conditioned multi-view 2D motion diffusion model is trained on the domain-specific synthetic data to recover 3D human motion and 3D HOI in the world space.
• Score: 41.002383749585924
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Reconstructing 3D human motion and human-object interactions (HOI) from Internet videos is a fundamental step toward building large-scale datasets of human behavior. Existing methods struggle to recover globally consistent 3D motion under dynamic cameras, especially for motion types underrepresented in current motion-capture datasets, and face additional difficulty recovering coherent human-object interactions in 3D. We introduce a two-stage framework leveraging 2D diffusion that reconstructs 3D human motion and HOI from Internet videos. In the first stage, we synthesize multi-view 2D motion data for each domain, leveraging 2D keypoints extracted from Internet videos to incorporate human motions that rarely appear in existing MoCap datasets. In the second stage, a camera-conditioned multi-view 2D motion diffusion model is trained on the domain-specific synthetic data to recover 3D human motion and 3D HOI in the world space. We demonstrate the effectiveness of our method on Internet videos featuring challenging motions such as gymnastics, as well as in-the-wild HOI videos, and show that it outperforms prior work in producing realistic human motion and human-object interaction.

Related papers

• PhyMotion: Structured 3D Motion Reward for Physics-Grounded Human Video Generation [75.96085587438279]
  Existing video rewards primarily rely on 2D perceptual signals, without explicitly modeling the 3D body state, contact, and dynamics underlying articulated human motion.<n>We propose PhyMotion, a structured, fine-grained motion reward that grounds recovered 3D human trajectories in a physics simulator and evaluates motion quality along multiple dimensions of physical feasibility.<n>Experiments show that PhyMotion achieves stronger correlation with human judgments than existing reward formulations.
  arXiv  Detail & Related papers  (2026-05-14T02:12:13Z)
• CoMoVi: Co-Generation of 3D Human Motions and Realistic Videos [34.06338037793912]
  CoMoVi is a co-generative framework that couples two video diffusion models (VDMs) to generate 3D human motions and videos synchronously within a single diffusion denoising loop.<n>In this paper, we propose an effective 2D human motion representation that can inherit the powerful prior of pre-trained VDMs.<n>We then design a dual-branch diffusion model to couple human motion and video generation process with mutual feature interaction and 3D-2D cross attentions.
  arXiv  Detail & Related papers  (2026-01-15T17:52:29Z)
• UniMo: Unifying 2D Video and 3D Human Motion with an Autoregressive Framework [54.337290937468175]
  We propose UniMo, an autoregressive model for joint modeling of 2D human videos and 3D human motions within a unified framework.<n>We show that our method simultaneously generates corresponding videos and motions while performing accurate motion capture.
  arXiv  Detail & Related papers  (2025-12-03T16:03:18Z)
• Mocap-2-to-3: Multi-view Lifting for Monocular Motion Recovery with 2D Pretraining [49.223455189395025]
  Mocap-2-to-3 is a novel framework that performs multi-view lifting from monocular input.<n>To leverage abundant 2D data, we decompose complex 3D motion into multi-view syntheses.<n>Our method surpasses state-of-the-art approaches in both camera-space motion realism and world-grounded human positioning.
  arXiv  Detail & Related papers  (2025-03-05T06:32:49Z)
• Motion-2-to-3: Leveraging 2D Motion Data to Boost 3D Motion Generation [43.915871360698546]
  2D human videos offer a vast and accessible source of motion data, covering a wider range of styles and activities.<n>We introduce a novel framework that disentangles local joint motion from global movements, enabling efficient learning of local motion priors from 2D data.<n>Our method efficiently utilizes 2D data, supporting realistic 3D human motion generation and broadening the range of motion types it supports.
  arXiv  Detail & Related papers  (2024-12-17T17:34:52Z)
• Lifting Motion to the 3D World via 2D Diffusion [19.64801640086107]
  We introduce MVLift, a novel approach to predict global 3D motion using only 2D pose sequences for training.<n> MVLift generalizes across various domains, including human poses, human-object interactions, and animal poses.
  arXiv  Detail & Related papers  (2024-11-27T23:26:56Z)
• EgoGaussian: Dynamic Scene Understanding from Egocentric Video with 3D Gaussian Splatting [95.44545809256473]
  EgoGaussian is a method capable of simultaneously reconstructing 3D scenes and dynamically tracking 3D object motion from RGB egocentric input alone. We show significant improvements in terms of both dynamic object and background reconstruction quality compared to the state-of-the-art.
  arXiv  Detail & Related papers  (2024-06-28T10:39:36Z)
• Learning Human Motion from Monocular Videos via Cross-Modal Manifold Alignment [45.74813582690906]
  Learning 3D human motion from 2D inputs is a fundamental task in the realms of computer vision and computer graphics. We present the Video-to-Motion Generator (VTM), which leverages motion priors through cross-modal latent feature space alignment. The VTM showcases state-of-the-art performance in reconstructing 3D human motion from monocular videos.
  arXiv  Detail & Related papers  (2024-04-15T06:38:09Z)
• Action2video: Generating Videos of Human 3D Actions [31.665831044217363]
  We aim to tackle the interesting yet challenging problem of generating videos of diverse and natural human motions from prescribed action categories. Key issue lies in the ability to synthesize multiple distinct motion sequences that are realistic in their visual appearances. Action2motionally generates plausible 3D pose sequences of a prescribed action category, which are processed and rendered by motion2video to form 2D videos.
  arXiv  Detail & Related papers  (2021-11-12T20:20:37Z)
• Estimating 3D Motion and Forces of Human-Object Interactions from Internet Videos [49.52070710518688]
  We introduce a method to reconstruct the 3D motion of a person interacting with an object from a single RGB video. Our method estimates the 3D poses of the person together with the object pose, the contact positions and the contact forces on the human body.
  arXiv  Detail & Related papers  (2021-11-02T13:40:18Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.