SCoPE
Sightline-Coordinate Positional Encoding for Video Diffusion Transformers
AI 导读
SCoPE 面向视频扩散Transformer的Sightline-坐标位置编码 模型：https://huggingface.co/TencentARC/SCoPE
SCoPE
Sightline-Coordinate Positional Encoding for Video Diffusion Transformers
来源：AK· x.com