Weakly Supervised Video Moment Retrieval From Text Queries
This addresses the scalability issue of acquiring temporal annotations for video retrieval, making it more practical for real-world applications.
The paper tackles the problem of text-to-video moment retrieval without requiring temporal boundary annotations during training, proposing a weakly supervised approach using only video-text pairs. Their method achieves comparable performance to state-of-the-art fully supervised approaches on two benchmark datasets.
There have been a few recent methods proposed in text to video moment retrieval using natural language queries, but requiring full supervision during training. However, acquiring a large number of training videos with temporal boundary annotations for each text description is extremely time-consuming and often not scalable. In order to cope with this issue, in this work, we introduce the problem of learning from weak labels for the task of text to video moment retrieval. The weak nature of the supervision is because, during training, we only have access to the video-text pairs rather than the temporal extent of the video to which different text descriptions relate. We propose a joint visual-semantic embedding based framework that learns the notion of relevant segments from video using only video-level sentence descriptions. Specifically, our main idea is to utilize latent alignment between video frames and sentence descriptions using Text-Guided Attention (TGA). TGA is then used during the test phase to retrieve relevant moments. Experiments on two benchmark datasets demonstrate that our method achieves comparable performance to state-of-the-art fully supervised approaches.