OpenAVS: Training-Free Open-Vocabulary Audio Visual Segmentation with Foundational Models

Open in new window