SOTAVerified

Weakly Supervised Action Localization

In this task, the training data consists of videos with a list of activities in them without any temporal boundary annotations. However, while testing, given a video, the algorithm should recognize the activities in the video and also provide the start and end time.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–10 of 55 papers

TitleStatusHype
Multilevel semantic and adaptive actionness learning for weakly supervised temporal action localizationCode0
Weakly supervised temporal action localization with actionness-guided false positive suppressionCode0
Revisiting Foreground and Background Separation in Weakly-supervised Temporal Action Localization: A Clustering-based ApproachCode1
Is Weakly-supervised Action Segmentation Ready For Human-Robot Interaction? No, Let's Improve It With Action-union LearningCode2
POTLoc: Pseudo-Label Oriented Transformer for Point-Supervised Temporal Action Localization—0
HR-Pro: Point-supervised Temporal Action Localization via Hierarchical Reliability PropagationCode1
Weakly-Supervised Action Localization by Hierarchically-structured Latent Attention Modeling—0
Actionness Inconsistency-guided Contrastive Learning for Weakly-supervised Temporal Action LocalizationCode1
Proposal-Based Multiple Instance Learning for Weakly-Supervised Temporal Action LocalizationCode1
Two-Stream Networks for Weakly-Supervised Temporal Action Localization With Semantic-Aware Mechanisms—0
Show:102550
← PrevPage 1 of 6Next →
All datasetsTHUMOS 2014THUMOS14ActivityNet-1.2ActivityNet-1.3GTEABEOIDFineActionTHUMOS' 14

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1ProCLavg-mAP (0.1:0.7)47.7—Unverified