The Solution for Temporal Sound Localisation Task of ICCV 1st Perception Test Challenge 2023
This work addresses the challenge of accurately localizing sounds in time for applications like video analysis, though it is incremental as it builds on existing multimodal and Transformer methods.
The paper tackled the problem of temporal sound localization by employing a multimodal fusion approach combining visual and audio features, achieving a mean average precision (mAP) of 0.33 and securing second-best performance in the ICCV 1st Perception Test Challenge 2023.
In this paper, we propose a solution for improving the quality of temporal sound localization. We employ a multimodal fusion approach to combine visual and audio features. High-quality visual features are extracted using a state-of-the-art self-supervised pre-training network, resulting in efficient video feature representations. At the same time, audio features serve as complementary information to help the model better localize the start and end of sounds. The fused features are trained in a multi-scale Transformer for training. In the final test dataset, we achieved a mean average precision (mAP) of 0.33, obtaining the second-best performance in this track.