How Well Do Self-Supervised Models Transfer? : A benchmark for evaluating self-supervision consisting of many-shot/few-shot recognition, object detection, surface normal estimation and semantic segmentation

How Well Do Self-Supervised Models Transfer? : A benchmark for evaluating self-supervision consisting of many-shot/few-shot recognition, object detection, surface normal estimation and semantic segmentation figure — AlphaXiv 中文概览（可滚动查看）