Interaction-Integrated Network for Natural Language Moment Localization
Ke Ning, Lingxi Xie, Jianzhuang Liu, Fei Wu, Qi Tian
Abstract
Natural language moment localization aims at localizing video clips according to a natural language description. The key to this challenging task lies in modeling the relationship between verbal descriptions and visual contents. Existing approaches often sample a number of clips from the video, and individually determine how each of them is related to the query sentence. However, this strategy can fail dramatically, in particular when the query sentence refers to some visual elements that appear outside of, or even are distant from, the target clip. In this paper, we address this issue by designing an Interaction-Integrated Network (I <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> N), which contains a few Interaction-Integrated Cells (I <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> Cs). The idea lies in the observation that the query sentence not only provides a description to the video clip, but also contains semantic cues on the structure of the entire video. Based on this, I <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> Cs go one step beyond modeling short-term contexts in the time domain by encoding long-term video content into every frame feature. By stacking a few I <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> Cs, the obtained network, I <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> N, enjoys an improved ability of inference, brought by both (I) multi-level correspondence between vision and language and (II) more accurate cross-modal alignment. When evaluated on a challenging video moment localization dataset named DiDeMo, I <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> N outperforms the state-of-the-art approach by a clear margin of 1.98%. On other two challenging datasets, Charades-STA and TACoS, I <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> N also reports competitive performance.