Hear-Your-Click: Interactive Video-to-Audio Generation via Object-aware Contrastive Audio-Visual Fine-tuning

Abstract: Video-to-audio (V2A) generation shows great potential in fields such as film production. Despite significant advances, current V2A methods, which rely on global video information, struggle with complex scenes and often fail to generate audio tailored to specific objects or regions in the videos. To address these limitations, we introduce Hear-Your-Click, an interactive V2A framework that enables users to generate sounds for specific objects in the videos by simply clicking on the frame. To achieve this, we propose Object-aware Contrastive Audio-Visual Fine-tuning (OCAV) with a Mask-guided Visual Encoder (MVE) to obtain object-level visual features aligned with corresponding audio segments. Furthermore, we tailor two data augmentation strategies: Random Video Stitching (RVS) and Mask-guided Loudness Modulation (MLM), aimed at enhancing the model's sensitivity to the segmented objects. To effectively measure the audio-visual correspondence, we design a new evaluation metric, the CAV score, for evaluation. Extensive experiments demonstrate that our framework offers more precise control and improved generation performance across various metrics. Project Page: https://github.com/SynapGrid/Hear-Your-Click