CVPR2021

Sparse R-CNN: End-to-End Object Detection With Learnable Proposals

Peize Sun, Rufeng Zhang, Yi Jiang, Tao Kong, Chenfeng Xu, Wei Zhan, Masayoshi Tomizuka, Lei Li, Zehuan Yuan, Changhu Wang, Ping Luo

DOI Publisher

Abstract

Figure 2 -Convergence curves of RetinaNet, Faster R-CNN, DETR and Sparse R-CNN on COCO val2017 [24]. Sparse R-CNN achieves competitive performance in terms of training efficiency and detection quality.