[Paper阅读] PointCLIP V2: Adapting CLIP for Powerful 3D Open-wor...
; GPT-3: “The obscure depth map shows a [table] at the corner.” 将四种描述集成为CLIP textual encoder的类别prompt. Application Zero-shot Classification. Few-shot Classification. Zero-shot Part Segmentation. Zero-shot 3D Object Detection....