AI Paper Atlas

Research map / Video / Action recognition

Adaptive RNN Tree for Large-Scale Human Action Recognition

Wenbo Li, Longyin Wen, Ming-Ching Chang, Ser Nam Lim, Siwei Lyu

ICCV 2017 · Video › Action recognition · approach: Skeleton-based action recognition · 125 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 early jump-out of recognized classes (ablation)HBRNN-L-T · 3D-SAR-140accuracy: 0.700 -> 0.756Improved
add inefficiency penalization (ablation)HBRNN-L-T · 3D-SAR-140accuracy: 0.697 -> 0.756Improved
add joint fine-tuning of RNNs (ablation)HBRNN-L-T · 3D-SAR-140accuracy: 0.733 -> 0.756Improved
increase tree levels from 1 to 2 (ablation)HBRNN-L-T · 3D-SAR-140accuracy: 0.604 -> 0.735Improved
increase tree levels from 2 to 3 (ablation)HBRNN-L-T · 3D-SAR-140accuracy: 0.735 -> 0.750Improved
increase tree levels from 3 to 4 (ablation)HBRNN-L-T · 3D-SAR-140accuracy: 0.750 -> 0.756Improved
allow overlapping action categories (theta_o 0.2 vs 1 which gives plain decision tree) (ablation)RNN-T/ACH · 3D-SAR-140accuracyImproved
wrap URNN-2L in RNN treeURNN-2L · 3D-SAR-140accuracy: 0.866 -> 0.892Improved
wrap HBRNN-L in RNN treeHBRNN-L · 3D-SAR-140accuracy: 0.604 -> 0.756Improved
wrap URNN in RNN treeURNN · 3D-SAR-140accuracy: 0.296 -> 0.539Improved
wrap URNN-2L in RNN treeURNN-2L · NTU RGB+D cross-viewaccuracy: 0.809 -> 0.832Improved
wrap URNN-2L in RNN treeURNN-2L · NTU RGB+D cross-subjectaccuracy: 0.730 -> 0.746Improved

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