AI Paper Atlas

Research map / Video / Video generation

Motion Mamba: Efficient and Long Sequence Motion Generation

Zeyu Zhang, Akide Liu, Ian Reid, RICHARD HARTLEY, Bohan Zhuang, Hao Tang

ECCV 2024 · Video › Video generation · approach: Text-to-motion diffusion · 109 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use hierarchical scan schedule {S_2N-1..S_1} instead of ascending scans {S_1..S_N} (ablation)Motion Mamba HTM block · HumanML3DFID: 1.278 -> 0.281Improved
Reverse the scan order to decrease with depth (ablation)Motion Mamba HTM block · HumanML3DFID: 1.278 -> 0.962Improved
Use block-level bidirectional scan in BSM instead of single scan (ablation)Motion Mamba BSM block · HumanML3DFID: 1.063 -> 0.281Improved
Use block-level instead of layer-level bidirectional scan (ablation)Motion Mamba BSM block · HumanML3DFID: 0.789 -> 0.281Improved
Set latent dimension to 2 instead of 1 (ablation)Motion Mamba · HumanML3DFID: 0.652 -> 0.281Improved
Increase latent dimension from 2 to 10 (ablation)Motion Mamba · HumanML3DFID: 0.281 -> 0.628Hurt
Increase depth from 11 to 27 or 37 layers (ablation)Motion Mamba · HumanML3DFID: 0.281 -> 0.975 (27 layers), 0.809 (37 layers)Hurt
Reduce depth from 11 to 9 layers (ablation)Motion Mamba · HumanML3DFID: 0.281 -> 1.080Hurt
Replace transformer diffusion denoiser with Mamba-based Motion MambaMLD · HumanML3Dinference time per sequence: 0.217 s -> 0.058 sImproved
Use Motion Mamba instead of MDM on diversity-style metricsMotion Mamba · HumanML3DMModality: 2.799 -> 2.294Hurt
Use Motion Mamba instead of MLD on long sequencesMotion Mamba · HumanML3D-LSFID: 0.952 -> 0.668Improved

Later papers that compared with this method

1 later accepted paper reports Motion Mamba as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy