Temporal Sentence Grounding in Streaming Videos

TL;DR

This paper introduces the task of Temporal Sentence Grounding in Streaming Videos, proposing novel methods to process continuous video streams on-the-fly for accurate query relevance evaluation.

Contribution

It presents two innovative techniques, TwinNet and a language-guided feature compressor, to address challenges in streaming video analysis for sentence grounding.

Findings

Outperforms existing methods on ActivityNet Captions, TACoS, and MAD datasets.

Demonstrates effectiveness of TwinNet and feature compressor through ablation studies.

Achieves real-time processing capabilities for streaming videos.

Abstract

This paper aims to tackle a novel task - Temporal Sentence Grounding in Streaming Videos (TSGSV). The goal of TSGSV is to evaluate the relevance between a video stream and a given sentence query. Unlike regular videos, streaming videos are acquired continuously from a particular source, and are always desired to be processed on-the-fly in many applications such as surveillance and live-stream analysis. Thus, TSGSV is challenging since it requires the model to infer without future frames and process long historical frames effectively, which is untouched in the early methods. To specifically address the above challenges, we propose two novel methods: (1) a TwinNet structure that enables the model to learn about upcoming events; and (2) a language-guided feature compressor that eliminates redundant visual frames and reinforces the frames that are relevant to the query. We conduct extensive experiments using ActivityNet Captions, TACoS, and MAD datasets. The results demonstrate the superiority of our proposed methods. A systematic ablation study also confirms their effectiveness.