Affinity Derivation for Accurate Instance Segmentation

doi:10.1145/3407090

Affinity Derivation for Accurate Instance Segmentation

2021-04-16

0

PRE

AI

DOI：10.1145/3407090

Original request for help

Save

Abstract

En 中文

Affinity, which represents whether two pixels belong to a same instance, is an equivalent representation to the instance segmentation labels. Conventional works do not make an explicit exploration on the affinity. In this article, we present two instance segmentation schemes based on pixel affinity information and show the effectiveness of affinity in both aspects. For proposal-free method, we predict pixel affinity for each image and then propose a simple yet effective graph merge algorithm to cluster pixels into instances. It shows that the affinity is powerful as an instance-relevant information to guide the clustering procedure in proposal-free instance segmentation. For proposal-based methods, we extend conventional framework with affinity head and introduce affinity as attached supervision in training phase. Without any additional inference cost, we can improve the performance of existing proposal-based instance segmentation methods, which shows that the affinity can also be applied as an auxiliary loss and training with such extra loss is beneficial to the training progress. Experimental results show that our schemes achieve comparable performance to other state-of-the-art instance segmentation methods. With Cityscapes training data, the proposed proposal-free method achieves 28.8 AP and the proposal-based method gets 27.2 AP both on test sets.

Keywords:

Instance segmentation

semantic segmentation

graph

Affinity Derivation for Accurate Instance Segmentation

Abstract

AI Summary

Journal

Organization