AI Paper Atlas

Research map / Video / Action recognition

Adversarial Background-Aware Loss for Weakly-supervised Temporal Activity Localization

Kyle Min, Jason J. Corso

ECCV 2020 · Video › Action recognition · approach: Weakly supervised temporal action localization · 113 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 the new background-vs-activity triplet (ablation)ATCL · THUMOS14mAP AVG: 27.4 -> 29.2Improved
add the new triplet (ablation)ATCL · THUMOS14mAP@IoU 0.5: 25.7 -> 28.9Improved
add an adversarial branch without the new triplet (ablation)ATCL · THUMOS14mAP AVG: 27.4 -> 27.2No clear effect
add an adversarial branch on top of the new triplet (ablation)ACL-PT · THUMOS14mAP AVG: 29.2 -> 30.0Improved
add an adversarial branch on top of the new triplet (ablation)ACL-PT · THUMOS14mAP@IoU 0.5: 28.9 -> 30.1Improved
use A2CL-PT instead of BaS-NetBaS-Net · THUMOS14mAP AVG: 27.9 -> 30.0Improved
use A2CL-PT instead of BaS-NetBaS-Net · ActivityNet1.3mAP AVG: 22.2 -> 22.5Improved
set adversarial TCAM weight omega greater than 1 (ablation)A2CL-PT · THUMOS14mAPHurt
use positive weight alpha on the A2CL-PT loss term (ablation)A2CL-PT · THUMOS14mAPImproved

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