AI Paper Atlas

Research map / Video / Action recognition

Spot On: Action Localization from Pointly-Supervised Proposals

Pascal Mettes, Jan C. van Gemert, Cees G. M. Snoek

ECCV 2016 · Video › Action recognition · approach: Weakly supervised temporal action localization · 103 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
train classifier on best-IoU proposals instead of ground-truth tubes (ablation)Action localization classifier (APT proposals) · UCF Sports and UCF 101localization AP across IoU thresholds: slightly improved accuracyNo clear effect
mine proposals from point annotations instead of using ground-truth boxes (ablation)MIL proposal mining · UCF Sports and UCF 101localization AP across IoU thresholds: competitive to the ground truth tubes across all thresholdsNo clear effect
use point supervision instead of video-level labels only in MILProposal mining · UCF SportsAUC (IoU 0.2): 0.278 -> 0.545Improved
use point supervision instead of video-level labels only in MILProposal mining · UCF 101mAP (IoU 0.2): 0.136 -> 0.348Improved
use point supervision instead of video-level labels only in MILProposal mining · Hollywood2TubesmAP (IoU 0.2): 0.009 -> 0.143Improved
point supervision versus box supervision (APT) for trainingThis work · UCF SportsAUC (IoU 0.2): 0.546 -> 0.545No clear effect
point supervision versus box supervision (APT) for trainingThis work · UCF 101mAP (IoU 0.2): 0.345 -> 0.348No clear effect
annotate points instead of boxes at every frameAnnotation process · UCF Sports and UCF 101annotation time: 10-15 times fasterImproved
annotate only 10% of frames with points (frame rate 10) (ablation)Point-based proposal mining · UCF Sports and UCF 101mAP at IoU 0.2 and 0.5: 45 times or more annotation speedupNo clear effect
annotate fewer frames (lower annotation frame-rate) (ablation)Point and box supervision · UCF Sports and UCF 101mAP: performance generally decreases as less frames are annotatedHurt
iterative proposal mining (10 iterations) guided by points (ablation)Proposal mining · UCF Sports (qualitative)IoU of top proposal with ground truth: 0.194 -> 0.627 and 0.401 -> 0.526Mixed

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