Exploiting Temporal Relationships in Video Moment Localization with Natural Language
arXiv:1908.03846
Abstract
We address the problem of video moment localization with natural language, i.e. localizing a video segment described by a natural language sentence. While most prior work focuses on grounding the query as a whole, temporal dependencies and reasoning between events within the text are not fully considered. In this paper, we propose a novel Temporal Compositional Modular Network (TCMN) where a tree attention network first automatically decomposes a sentence into three descriptions with respect to the main event, context event and temporal signal. Two modules are then utilized to measure the visual similarity and location similarity between each segment and the decomposed descriptions. Moreover, since the main event and context event may rely on different modalities (RGB or optical flow), we use late fusion to form an ensemble of four models, where each model is independently trained by one combination of the visual input. Experiments show that our model outperforms the state-of-the-art methods on the TEMPO dataset.
References in corpus (6)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Temporal Segment Networks: Towards Good Practices for Deep Action Recognition
- TALL: Temporal Activity Localization via Language Query
- MAttNet: Modular Attention Network for Referring Expression Comprehension
- Constituency Parsing with a Self-Attentive Encoder
- Improving Referring Expression Grounding with Cross-modal Attention-guided Erasing