AI Paper Atlas

Research map / Video / Action recognition

Human Action Recognition From Various Data Modalities: A Review

Zehua Sun, Qiuhong Ke, Hossein Rahmani, Mohammed Bennamoun, Gang Wang, Jun Liu

TPAMI 2022 · Video › Action recognition · approach: Disentangled action recognition · 733 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
move from skeleton RNN (Trust Gate ST-LSTM) to graph convolution (InfoGCN)skeleton HAR · NTU RGB+D (cross-subject)accuracy: 69.2 -> 93.0Improved
move from ST-GCN to CTR-GCNskeleton GCN · NTU RGB+D (cross-subject)accuracy: 81.5 -> 92.4Improved
add Transformer attention to skeleton models (ST-TR vs ST-GCN style)skeleton HAR · NTU RGB+D (cross-subject)accuracy: 89.9 -> 92.3Mixed
move from NL I3D to Transformer MTV-HRGB video HAR · Kinetics-400accuracy: 77.7 -> 89.1Improved
use 3D CNN with BERT head instead of two-stream 2D CNN (TSN)RGB video HAR · UCF101accuracy: 94.2 -> 98.7Improved
use RNN-based DB-LSTM instead of Soft Attention ModelRGB video HAR · HMDB51accuracy: 41.3 -> 81.2Improved

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