arXiv Analytics

Sign in

arXiv:1705.01861 [cs.CV]AbstractReferencesReviewsResources

Action Tubelet Detector for Spatio-Temporal Action Localization

Vicky Kalogeiton, Philippe Weinzaepfel, Vittorio Ferrari, Cordelia Schmid

Published 2017-05-04Version 1

Current state-of-the-art approaches for spatio-temporal action detection rely on detections at the frame level that are then linked or tracked across time. In this paper, we leverage the temporal continuity of videos instead of operating at the frame level. We propose the ACtion Tubelet detector (ACT-detector) that takes as input a sequence of frames and outputs tubeletes, i.e., sequences of bounding boxes with associated scores. The same way state-of-the-art object detectors rely on anchor boxes, our ACT-detector is based on anchor cuboids. We build upon the state-of-the-art SSD framework (Single Shot MultiBox Detector). Convolutional features are extracted for each frame, while scores and regressions are based on the temporal stacking of these features, thus exploiting information from a sequence. Our experimental results show that leveraging sequences of frames significantly improves detection performance over using individual frames. The gain of our tubelet detector can be explained by both more relevant scores and more precise localization. Our ACT-detector outperforms the state of the art methods for frame-mAP and video-mAP on the J-HMDB and UCF-101 datasets, in particular at high overlap thresholds.

Related articles: Most relevant | Search more
arXiv:2106.08061 [cs.CV] (Published 2021-06-15)
Relation Modeling in Spatio-Temporal Action Localization
Yutong Feng et al.
arXiv:1506.01929 [cs.CV] (Published 2015-06-05)
Learning to track for spatio-temporal action localization
arXiv:1905.02171 [cs.CV] (Published 2019-05-06)
Spatio-Temporal Action Localization in a Weakly Supervised Setting