MELON: A Large-Scale Dataset for Multi-Event Text-to-Long-Video Retrieval

2026-08-31

AI Summary

为解决长视频多事件文本检索问题，构建了MELON数据集，并提出了一种多事件感知损失函数以提高检索精度。

Abstract

Existing text-video retrieval datasets primarily consist of short-form clips containing a single dominant event. While suitable for measuring basic vision-language alignment, they are limited in capturing real-world retrieval scenarios, where long-form videos naturally contain multiple semantically distinct events and a single text query may correspond to several non-contiguous temporal segments. To bridge this gap, we introduce MELON, the first large-scale dataset designed to extend text-video retrieval to long-form videos featuring complex, multi-event structures. MELON explicitly annotates multiple event intervals per video along with their corresponding textual descriptions, enabling both training and evaluation of multi-event understanding in long, untrimmed videos. In addition, we propose a multi-event aware loss that encourages models to differentiate between full-event and partial-event matches, yielding substantial improvements in retrieval accuracy. Together, the MELON dataset and our proposed loss establish a robust foundation for expanding text-to-video retrieval to complex long-form scenarios and provide a more realistic evaluation setting for future research in the field.

Problem

• text-video retrieval
• long-form videos
• multi-event structures
• real-world scenarios

Innovation

• multi-event text-to-long-video retrieval
• MELON dataset
• multiple event intervals
• multi-event aware loss
• full-event and partial-event matches