Visual-Semantic Understanding

PhD thesis investigating critical challenges associated with multimodal attentive architectures, including improving semantic segmentation accuracy, enabling open-vocabulary segmentation, and advancing video question answering.