Learning to Cluster for Proposal-Free Instance Segmentation

17 Mar 2018Yen-Chang HsuZheng XuZsolt KiraJiawei Huang

This work proposed a novel learning objective to train a deep neural network to perform end-to-end image pixel clustering. We applied the approach to instance segmentation, which is at the intersection of image semantic segmentation and object detection... (read more)

PDF Abstract

Evaluation Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK COMPARE
Lane Detection TuSimple Pairwise pixel supervision + FCN Accuracy 96.50% # 3