AI Paper Atlas

Research map / Video / Action recognition

Video Action Recognition with Attentive Semantic Units

Yifei Chen, Dapeng Chen, Ruijin Liu, Hao Li, Wei Peng

ICCV 2023 · Video › Action recognition · approach: Disentangled action recognition · 22 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
replace transformer with semantic-unit-queried cross-attention decoder (ablation)ASU-B/16 · Kinetics-400top-1 accuracy: 81.1 -> 82.0Improved
add region-level representation (MRA) (ablation)ASU-B/16 · Kinetics-400top-1 accuracy: 81.1 -> 81.8Improved
combine region-level representation with semantic units (ablation)ASU-B/16 · Kinetics-400top-1 accuracy: 82.0 -> 82.4Improved
add 1D temporal convolution in video decoder (ablation)ASU-B/16 · Kinetics-400top-1 accuracy: 82.4 -> 82.6Improved
add semantic units as decoder queries (ablation)ASU-B/16 · HMDB-51 2-shottop-1 accuracy: 52.6 -> 57.7Improved
add semantic units as decoder queries (ablation)ASU-B/16 · UCF-101 zero-shottop-1 accuracy: 64.9 -> 75.0Improved
use only one type of semantic unit instead of all four (ablation)ASU-B/16 · UCF-101 zero-shottop-1 accuracy: 75.0 -> 73.8 (objects only)Hurt
increase regions from 4 to 8 using 4x2 split (ablation)ASU-B/16 · Kinetics-400top-1 accuracy: 82.6 -> 81.8Hurt
use vertical divisions of regions (2x2) (ablation)ASU-B/16 · Kinetics-400top-1 accuracy: 82.6 -> 82.2Hurt
use learnable prompts (CoOp, CoCoOp) instead of fixed template (ablation)ASU-B/16 · Kinetics-400top-1 accuracy: 82.6 -> 82.3 (CoOp), 82.2 (CoCoOp)No clear effect
apply ASU in zero-shot on HMDB-51 and UCF-101 vs X-CLIP-B/16X-CLIP-B/16 · HMDB-51 / UCF-101 zero-shottop-1 accuracy: 44.6 -> 48.1; 72.0 -> 75.0Improved

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