AI Paper Atlas

Research map / Video / Action recognition

Learning Action Completeness From Points for Weakly-Supervised Temporal Action Localization

Pilhyeon Lee, Hyeran Byun

ICCV 2021 · Video › Action recognition · approach: Weakly supervised temporal action localization · 81 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 point-level supervision loss (ablation)video-level baseline · THUMOS'14AVG mAP 0.1:0.7: 28.7 -> 47.3Improved
add score contrastive loss (ablation)baseline with point loss · THUMOS'14AVG mAP 0.1:0.7: 47.3 -> 52.0Improved
add feature contrastive loss (ablation)baseline with point loss · THUMOS'14AVG mAP 0.1:0.7: 47.3 -> 49.0Improved
add feature contrastive loss on top of score loss (ablation)score-loss model · THUMOS'14AVG mAP 0.1:0.7: 52.0 -> 52.8Improved
score sequences by action contrast instead of inner scores (ablation)optimal sequence search · THUMOS'14AVG mAP 0.1:0.7: 49.0 -> 50.8Improved
include background instances in score contrast (ablation)optimal sequence search · THUMOS'14AVG mAP 0.1:0.7: 50.8 -> 52.8Improved
increase greedy search budget from 1 to 25 (ablation)optimal sequence search · THUMOS'14AVG mAP 0.1:0.7: 51.3 -> 52.8Improved
increase greedy search budget beyond 5 (ablation)optimal sequence search · THUMOS'14AVG mAP 0.1:0.7: 52.5 -> 52.8No clear effect
use at-least-one background point per section instead of global mining (ablation)baseline without completeness learning · THUMOS'14AVG mAP 0.1:0.7: 43.6 -> 46.6Improved
add filling between background points (ablation)pseudo background mining · THUMOS'14AVG mAP 0.1:0.7: 46.6 -> 47.3Improved
use uniformly distributed point labels instead of Gaussian (ablation)Ours · THUMOS'14AVG mAP 0.1:0.7: 52.8 -> 49.3Hurt
use point-level labels instead of video-level labelsOurs vs Lee et al. · ActivityNet 1.3AVG mAP: 23.7 -> 25.1Improved

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