Weakly Supervised Video Representation Learning with Unaligned Text for Sequential Videos

TL;DR

This paper introduces a weakly supervised method for sequential video understanding that leverages unaligned text and pseudo labels, employing contrastive losses and a transformer-based video encoder to improve text-video matching.

Contribution

It proposes a novel weakly supervised framework using pseudo frame-sentence correspondence and multiple granularity contrastive losses for sequential video understanding.

Findings

Outperforms baseline methods significantly in video sequence verification.

Effective in text-to-video matching tasks with unaligned text.

Validates the use of pseudo labels for weakly supervised learning.

Abstract

Sequential video understanding, as an emerging video understanding task, has driven lots of researchers' attention because of its goal-oriented nature. This paper studies weakly supervised sequential video understanding where the accurate time-stamp level text-video alignment is not provided. We solve this task by borrowing ideas from CLIP. Specifically, we use a transformer to aggregate frame-level features for video representation and use a pre-trained text encoder to encode the texts corresponding to each action and the whole video, respectively. To model the correspondence between text and video, we propose a multiple granularity loss, where the video-paragraph contrastive loss enforces matching between the whole video and the complete script, and a fine-grained frame-sentence contrastive loss enforces the matching between each action and its description. As the frame-sentence correspondence is not available, we propose to use the fact that video actions happen sequentially in the temporal domain to generate pseudo frame-sentence correspondence and supervise the network training with the pseudo labels. Extensive experiments on video sequence verification and text-to-video matching show that our method outperforms baselines by a large margin, which validates the effectiveness of our proposed approach. Code is available at https://github.com/svip-lab/WeakSVR