Audio Visual Retrieval

Audio-visual retrieval focuses on developing systems that can effectively link audio and visual information, enabling tasks like searching for images based on audio descriptions or vice-versa. Current research emphasizes improving the accuracy and robustness of these systems, particularly by addressing limitations in handling negative audio examples and incorporating fine-grained object details. This involves exploring various model architectures, including contrastive learning, deep metric learning, and unified frameworks that integrate audio, visual, and textual information, often leveraging pre-trained models like CLIP and HuBERT. Advances in this field have significant implications for applications such as multimedia search, content creation, and assistive technologies for visually or hearing-impaired individuals.