English

W-TALC: Weakly-supervised Temporal Activity Localization and Classification

Computer Vision and Pattern Recognition 2018-12-18 v3

Abstract

Most activity localization methods in the literature suffer from the burden of frame-wise annotation requirement. Learning from weak labels may be a potential solution towards reducing such manual labeling effort. Recent years have witnessed a substantial influx of tagged videos on the Internet, which can serve as a rich source of weakly-supervised training data. Specifically, the correlations between videos with similar tags can be utilized to temporally localize the activities. Towards this goal, we present W-TALC, a Weakly-supervised Temporal Activity Localization and Classification framework using only video-level labels. The proposed network can be divided into two sub-networks, namely the Two-Stream based feature extractor network and a weakly-supervised module, which we learn by optimizing two complimentary loss functions. Qualitative and quantitative results on two challenging datasets - Thumos14 and ActivityNet1.2, demonstrate that the proposed method is able to detect activities at a fine granularity and achieve better performance than current state-of-the-art methods.

Keywords

Cite

@article{arxiv.1807.10418,
  title  = {W-TALC: Weakly-supervised Temporal Activity Localization and Classification},
  author = {Sujoy Paul and Sourya Roy and Amit K Roy-Chowdhury},
  journal= {arXiv preprint arXiv:1807.10418},
  year   = {2018}
}

Comments

Accepted at European Conference on Computer Vision (ECCV), 2018

R2 v1 2026-06-23T03:16:16.420Z