Kehan Li

semantic segmentation (3) diffusion model (3) temporal modeling (2) vision transformer (2) interactive segmentation (2) self-supervised learning (2) contrastive loss (2) video instance segmentation (2) data augmentation (1) weakly supervised learning (1) representation learning (1) prototype learning (1) open-vocabulary detection (1) model evaluation (1) contrastive learning (1) visual grounding (1) object detection (1) feature embedding (1) hierarchical representation (1) multimodal learning (1)