PointRend: Image Segmentation as Rendering

CVPR 2020 Alexander KirillovYuxin WuKaiming HeRoss Girshick

We present a new method for efficient high-quality image segmentation of objects and scenes. By analogizing classical computer graphics methods for efficient rendering with over- and undersampling challenges faced in pixel labeling tasks, we develop a unique perspective of image segmentation as a rendering problem... (read more)

PDF Abstract

Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT LEADERBOARD
Semantic Segmentation Cityscapes val SemanticFPN P2-P5 + PointRend mIoU 78.6% # 10

Methods used in the Paper