AI Paper Atlas

Research map / Video / Action recognition

Two-Stream Consensus Network for Weakly-Supervised Temporal Action Localization

Yuanhao Zhai, Le Wang, Wei Tang, Qilin Zhang, Junsong Yuan, Gang Hua

ECCV 2020 · Video › Action recognition · approach: Weakly supervised temporal action localization · 158 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 hard pseudo ground truth from fused attention (fusion) (ablation)TSCN with UntrimmedNet · THUMOS14 testmAP@IoU 0.5: 24.0 -> 27.6Improved
add hard pseudo ground truth to RGB stream (ablation)RGB stream · THUMOS14 testmAP@IoU 0.5: 8.2 -> 14.4Improved
add hard pseudo ground truth to flow stream (ablation)flow stream · THUMOS14 testmAP@IoU 0.5: 23.2 -> 24.1Improved
add hard pseudo ground truth to flow stream (recall) (ablation)flow stream · THUMOS14 testrecall at IoU 0.5: 43.3 -> 42.4Hurt
add hard pseudo ground truth to fusion (precision) (ablation)TSCN fusion · THUMOS14 testprecision at IoU 0.5: 23.6 -> 31.3Improved
use hard instead of soft pseudo ground truth (ablation)TSCN · THUMOS14 testmAPImproved
add attention normalization loss L_att instead of background classification loss L_bg (ablation)TSCN · THUMOS14 testmAPImproved
add background classification loss on top of L_att (ablation)TSCN · THUMOS14 testmAPHurt
switch features from UntrimmedNet to I3DTSCN · THUMOS14 testmAP@IoU 0.5: 27.6 -> 28.7Improved
TSCN versus BaS-Net with I3D features at high IoUTSCN · THUMOS14 testmAP@IoU 0.7: 10.4 -> 10.2Hurt
TSCN versus Nguyen et al. on ActivityNet v1.3 at IoU 0.5TSCN · ActivityNet v1.3 valmAP@IoU 0.5: 36.4 -> 35.3Hurt
TSCN versus Liu et al. on ActivityNet v1.2TSCN · ActivityNet v1.2 valaverage mAP: 22.4 -> 23.6Improved

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