AI Paper Atlas

Research map / Video / Action recognition

Amplify, Aggregate, and Adjust: VideoMAE-based Holistic-Subtle Aggregation for Micro-Action Recognition

Yan Zhang, Nan Pu, Wenjing Li, Zhun Zhong, Meng Wang

ECCV 2026 · Video › Action recognition · approach: Fully supervised temporal action detection · 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
add TGLA subtle-motion branch to VideoMAE baseline (ablation)VideoMAE baseline · MA-52Action Top-1: 66.47 -> 67.04Improved
add HSMA dual cross-attention on top of TGLA (ablation)VideoMAE + TGLA · MA-52Action Top-1: 67.04 -> 67.63Improved
add CADA on top of TGLA + HSMA (ablation)VideoMAE + TGLA + HSMA · MA-52Action Top-1: 67.63 -> 67.92Improved
add TGLA to baseline (macro F1) (ablation)VideoMAE baseline · MA-52Action F1 macro: 52.46 -> 55.09Improved
use only selective confidence distillation L_SCD in CADA (ablation)A3-MAE without CADA · MA-52Action Top-1: 67.63 -> 67.49Hurt
add confidence-weight alignment L_CWA to L_SCD (ablation)CADA · MA-52Action Top-1: 67.49 -> 67.60Improved
add weight entropy regularization L_WER to L_SCD (ablation)CADA · MA-52Action Top-1: 67.49 -> 67.69Improved
use all three CADA losses instead of L_SCD + L_WER (ablation)CADA · MA-52Action Top-1: 67.69 -> 67.92Improved
use aggregated head Ga instead of single-branch head (ablation)A3-MAE · MA-52Action Top-1: 67.60 -> 67.92Improved
use instance-adaptive CADA weights instead of fixed subtle-branch weight (ablation)A3-MAE · MA-52Action Top-1: every fixed setting underperforms CADA (67.92)Improved
use 16 frames instead of fewer frames per video (ablation)A3-MAE · MA-52Action Top-1: best results achieved at N = 16Improved
raise TG-mask threshold from 90 to larger values (ablation)A3-MAE · MA-52Action Top-1: 67.92 -> 67.08Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 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