Gensheng Pei

representation learning (1) contrastive learning (1) change detection (1) object detection (1) self-supervised learning (1) video understanding (1) remote sensing (1) edge detection (1) convolutional neural network (1) convolutional network (1) synthetic datum (1) vision-language model (1) image-text alignment (1) masked autoencoder (1) zero-shot classification (1) semantic preservation (1) hierarchical feature (1) layout-to-image generation (1) video representation (1) dual learning (1)