AI Paper Atlas

Research map / Video / Video generation

ARMS: Anchor–Relational Motion Streaming for Seamless Solo-Social Motion Transitions

Huakun Liu, Qing Yu, Kent Fujiwara, Hideaki Uchiyama, Kiyoshi Kiyokawa

ECCV 2026 · Video › Video generation · approach: Text-to-motion diffusion · 0 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 partner-referenced Anchor-Relational representation instead of InterGen global representation (ablation)ARMS · InterHuman transitionsAUJ: 3.367 -> 2.925Improved
use Anchor-Relational representation instead of InterGen global representation (ablation)ARMS · InterHuman interaction generationR-Precision Top-3: 0.723 -> 0.723; FID 4.553 -> 4.446No clear effect
add autoregressive streaming (ablation)ARMS · InterHumanR-Precision Top-3 / FID: 0.764 -> 0.723; FID 4.436 -> 4.446Hurt
add autoregressive streaming (ablation)ARMS · InterHuman transitionsAUJ: 3.209 -> 2.925Improved
use segment size 5 instead of 1 (ablation)ARMS · InterHumanFID: 6.832 -> 4.446Improved
use refinement offset 5 instead of 50 (ablation)ARMS · InterHumanFID: 8.645 -> 4.446Improved
use latent dim 64 instead of 128 (ablation)ARMS · InterHumanFID: 4.969 -> 4.446Improved
use latent dim 64 instead of 32 (ablation)ARMS · InterHumanAUJ: 2.894 -> 2.925; PJ 0.115 -> 0.071Mixed
increase sampling steps K from 10 to 50 (ablation)ARMS · InterHumanFID: 4.731 -> 4.446Improved
replace adapted InterMask with ARMS streamingARMS · Solo-to-interaction transitionPJ: 3.131 -> 0.077Improved
replace InterMask (Adapted) with ARMSARMS · Interaction-to-interaction streamingAUJ: 22.440 -> 2.925Improved
replace InterMask with ARMS streaming on subsequence semanticsARMS · Interaction subsequenceR@Top3: 0.557 -> 0.509Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 3 on most shared results.

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