Huisheng Wang

dense prediction (2) depth estimation (2) image generation (1) scene understanding (1) self-supervised learning (1) weakly supervised learning (1) text-to-image synthesis (1) data augmentation (1) multimodal learning (1) multi-label classification (1) egocentric vision (1) video understanding (1) diffusion model (1) supervised fine-tuning (1) classifier-free guidance (1) clip model (1) egocentric video (1) panoptic segmentation (1) subject-driven generation (1) prompt alignment (1)