AI Paper Atlas

Research map / Video / Action recognition

Revisiting Foreground and Background Separation in Weakly-supervised Temporal Action Localization: A Clustering-based Approach

Qinying Liu, Zilei Wang, Shenghai Rong, Junjie Li, Yixin Zhang

ICCV 2023 · Video › Action recognition · approach: Weakly supervised temporal action localization · 25 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 snippet clustering component (SCC) joint training (ablation)WTAL baseline · THUMOS14AVG mAP: 42.1 -> 43.2Improved
add cluster classification component (CCC) (ablation)baseline + SCC · THUMOS14AVG mAP: 43.2 -> 43.9Improved
add clustering-assisted testing (ablation)CASE · THUMOS14AVG mAP: 43.9 -> 46.2Improved
use optimal transport marginal constraint to generate pseudo-labels (ablation)SCC · THUMOS14AVG mAP: 40.8 -> 44.7 (hard); 41.2 -> 46.2 (soft)Improved
use soft instead of hard OT labels (ablation)SCC · THUMOS14AVG mAP: 44.7 -> 46.2Improved
use K=2 clusters instead of many clusters (ablation)CASE · THUMOS14AVG mAPHurt
increase K beyond 16 (ablation)CASE · THUMOS14AVG mAP: stable within +/-0.2%No clear effect
add Gaussian prior distribution (KL term) to SCC labeling (ablation)SCC · THUMOS14AVG mAP: 45.4 -> 46.2Improved
estimate CCC marginal from snippet foreground labels instead of equipartition (ablation)CCC · THUMOS14AVG mAP: 44.0 -> 46.2Improved
fuse attention and cluster-based foreground probabilities at test time (ablation)CASE · THUMOS14AVG mAP: 43.9 / 44.1 -> 46.2Improved
add CASE to other WTAL baselinesCO2-Net, AICL, Zhou et al. · THUMOS14AVG mAP: 44.6 -> 46.6; 46.4 -> 48.0; 48.3 -> 49.2Improved

Earlier methods this paper compares with

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