Papers › Motion Anything: Any to Motion Generation

Motion Anything: Any to Motion Generation

10 Mar 2025arXiv:2503.06955archive 2025-07-28

Zeyu Zhang, Yiran Wang, Wei Mao, Danning Li, Rui Zhao, Biao Wu, Zirui Song, Bohan Zhuang, Ian Reid, Richard Hartley

Conditional motion generation has been extensively studied in computer vision, yet two critical challenges remain. First, while masked autoregressive methods have recently outperformed diffusion-based approaches, existing masking models lack a mechanism to prioritize dynamic frames and body parts based on given conditions. Second, existing methods for different conditioning modalities often fail to integrate multiple modalities effectively, limiting control and coherence in generated motion. To address these challenges, we propose Motion Anything, a multimodal motion generation framework that introduces an Attention-based Mask Modeling approach, enabling fine-grained spatial and temporal control over key frames and actions. Our model adaptively encodes multimodal conditions, including text and music, improving controllability. Additionally, we introduce Text-Motion-Dance (TMD), a new motion dataset consisting of 2,153 pairs of text, music, and dance, making it twice the size of AIST++, thereby filling a critical gap in the community. Extensive experiments demonstrate that Motion Anything surpasses state-of-the-art methods across multiple benchmarks, achieving a 15% improvement in FID on HumanML3D and showing consistent performance gains on AIST++ and TMD. See our project website https://steve-zeyu-zhang.github.io/MotionAnything

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

steve-zeyu-zhang/MotionAnything officialmentioned on GitHub report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Motion GenerationMotion Synthesis

Datasets

Introduced by this paper, per the archive.

TMD

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Motion Synthesis AIST++ Motion Anything Beat alignment score 0.2757 #1 of 12 Archive leaderboard report
Motion Synthesis AIST++ Motion Anything FID 17.22 #1 of 12 Archive leaderboard report
Motion Synthesis HumanML3D Motion Anything Diversity 9.521 #1 of 37 Archive leaderboard report
Motion Synthesis HumanML3D Motion Anything FID 0.028 #1 of 37 Archive leaderboard report
Motion Synthesis HumanML3D Motion Anything Multimodality 2.705 #1 of 37 Archive leaderboard report
Motion Synthesis HumanML3D Motion Anything R Precision Top3 0.829 #1 of 37 Archive leaderboard report
Motion Synthesis KIT Motion-Language Motion Anything Diversity 10.94 #1 of 31 Archive leaderboard report
Motion Synthesis KIT Motion-Language Motion Anything FID 0.131 #1 of 31 Archive leaderboard report
Motion Synthesis KIT Motion-Language Motion Anything Multimodality 1.374 #1 of 31 Archive leaderboard report
Motion Synthesis KIT Motion-Language Motion Anything R Precision Top3 0.802 #1 of 31 Archive leaderboard report
Motion Synthesis TMD Motion Anything BAS 0.2094 #1 of 1 Archive leaderboard report
Motion Synthesis TMD Motion Anything FID 21.46 #1 of 1 Archive leaderboard report
Motion Synthesis TMD Motion Anything MMDist 5.34 #1 of 1 Archive leaderboard report
Motion Synthesis TMD Motion Anything MModality 2.424 #1 of 1 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections