VMM: Video-Music Mamba for generating background music from videos

Jiajun Xu, Zixiang Lu, Ping Gao, Qiguang Miao, Kun Xie

Published: 01 Dec 2025, Last Modified: 06 Nov 2025Computer Vision and Image UnderstandingEveryoneRevisionsCC BY-SA 4.0
Abstract: Highlights•First to integrate Mamba and Transformer for video music generation.•Switch aligns chord and video features via random input toggling and gradient control.•VMM outperforms SOTA in objective metrics for high-quality background music generation.
Loading