TR-DETR: Task-Reciprocal Transformer for Joint Moment Retrieval and Highlight Detection

TL;DR

TR-DETR introduces a novel task-reciprocal transformer that leverages the inherent relationship between video moment retrieval and highlight detection, improving performance through shared feature alignment and task cooperation.

Contribution

The paper proposes a task-reciprocal transformer that explicitly models the mutual influence between MR and HD, enhancing joint video analysis beyond existing separate or loosely coupled methods.

Findings

Outperforms state-of-the-art methods on multiple datasets

Effectively aligns multi-modal features into a shared space

Utilizes reciprocity to refine retrieval and highlight prediction

Abstract

Video moment retrieval (MR) and highlight detection (HD) based on natural language queries are two highly related tasks, which aim to obtain relevant moments within videos and highlight scores of each video clip. Recently, several methods have been devoted to building DETR-based networks to solve both MR and HD jointly. These methods simply add two separate task heads after multi-modal feature extraction and feature interaction, achieving good performance. Nevertheless, these approaches underutilize the reciprocal relationship between two tasks. In this paper, we propose a task-reciprocal transformer based on DETR (TR-DETR) that focuses on exploring the inherent reciprocity between MR and HD. Specifically, a local-global multi-modal alignment module is first built to align features from diverse modalities into a shared latent space. Subsequently, a visual feature refinement is designed to eliminate query-irrelevant information from visual features for modal interaction. Finally, a task cooperation module is constructed to refine the retrieval pipeline and the highlight score prediction process by utilizing the reciprocity between MR and HD. Comprehensive experiments on QVHighlights, Charades-STA and TVSum datasets demonstrate that TR-DETR outperforms existing state-of-the-art methods. Codes are available at \url{https://github.com/mingyao1120/TR-DETR}.