Roberto Amoroso

semantic segmentation (3) diffusion model (2) self-supervised learning (1) model evaluation (1) multimodal learning (1) false positive (1) open-vocabulary segmentation (1) vision-language model (1) text-conditioned generation (1) image-text alignment (1) video question answering (1) open vocabulary (1) training-free method (1) prototype generation (1) intersection over union (1) error analysis (1) false negative (1) visual embedding (1) large language model (1) temporal modeling (1)