AutoLoc: Weakly-supervised Temporal Action Localization in Untrimmed Videos

Zheng Shou, Hang Gao, Lei Zhang, Kazuyuki Miyazawa, Shih-Fu Chang; The European Conference on Computer Vision (ECCV), 2018, pp. 154-171

Abstract


Temporal Action Localization (TAL) in untrimmed video is important for many applications. But it is very expensive to annotate the segment-level ground truth (action class and temporal boundary). This raises the interest of addressing TAL with weak supervision, namely only video-level annotations are available during training). However, the state-of-the-art weakly-supervised TAL methods only focus on generating good Class Activation Sequence (CAS) over time but conduct simple thresholding on CAS to localize actions. In this paper, we first develop a novel weakly-supervised TAL framework called AutoLoc to directly predict the temporal boundary of each action instance. We propose a novel Outer-Inner-Contrastive (OIC) loss to automatically discover the needed segment-level supervision for training such a boundary predictor. Our method achieves dramatically improved performance: under the IoU threshold 0.5, our method improves mAP on THUMOS'14 from 13.7% to 21.2% and mAP on ActivityNet from 7.4% to 27.3%. It is also very encouraging to see that our weakly-supervised method achieves comparable results with some fully-supervised methods.

Related Material


[pdf]
[bibtex]
@InProceedings{Shou_2018_ECCV,
author = {Shou, Zheng and Gao, Hang and Zhang, Lei and Miyazawa, Kazuyuki and Chang, Shih-Fu},
title = {AutoLoc: Weakly-supervised Temporal Action Localization in Untrimmed Videos},
booktitle = {The European Conference on Computer Vision (ECCV)},
month = {September},
year = {2018}
}