AI Paper Atlas

Research map / Video / Action recognition

Learning Temporal Action Proposals With Fewer Labels

Jingwei Ji, Kaidi Cao, Juan Carlos Niebles

ICCV 2019 · Video › Action recognition · approach: Fully supervised temporal action detection · 47 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 Time Warping perturbation (ablation)Mean Teacher with Time Masking · THUMOS14 (60% labels)AR@100: 44.79 -> 46.87Improved
add Time Masking perturbation (ablation)Mean Teacher with Time Warping · THUMOS14 (60% labels)AR@100: 45.37 -> 46.87Improved
add Time Masking instead of regular dropout (ablation)Mean Teacher · THUMOS14 (60% labels)AR@100: 44.25 -> 46.87Improved
add regular dropout to Mean Teacher (ablation)Mean Teacher · THUMOS14 (60% labels)AR@100: 44.20 -> 44.25No clear effect
add VAT to Mean Teacher (ablation)Mean Teacher · THUMOS14 (60% labels)AR@100: 44.20 -> 44.21No clear effect
add Mean Teacher semi-supervised training with unlabeled videosBSN · THUMOS14 (60% labels)AR@100: 40.79 -> 44.20Improved
add VAT adversarial noiseBSN · THUMOS14 (60% labels)AR@100: 40.79 -> 43.13Improved
use semi-supervised framework with 60% labels instead of fully supervised BSN with 60% labelsBSN · THUMOS14AR@100: 40.79 -> 46.87Improved
semi-supervised with 60% labels vs BSN with all labelsBSN · ActivityNet v1.3AR@100: 74.16 -> 75.07Improved
semi-supervised with 60% labels vs BSN with all labelsBSN · THUMOS14AR@100: 46.06 -> 46.87Improved
add unlabeled-data training when 100% labels availableBSN · THUMOS14AR@100: 46.06 -> 47.53Improved
use semi-supervised proposals for localizationBSN + UNet · THUMOS14 testmAP@0.7: 20.0 -> 20.7Improved

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