Bridging Category-level and Instance-level Semantic Image Segmentation

TL;DR

This paper introduces a novel approach to instance-level image segmentation that builds on category-level segmentation, using a regression network to predict instance bounding boxes for each pixel, achieving state-of-the-art results.

Contribution

It presents a simple, effective method for semantic instance segmentation, including an online bootstrapping training technique and leveraging high-accuracy semantic segmentation models.

Findings

Achieved 79.1% mIoU on PASCAL VOC 2012

State-of-the-art results for instance segmentation

Effective alternative to detect-then-segment pipelines

Abstract

We propose an approach to instance-level image segmentation that is built on top of category-level segmentation. Specifically, for each pixel in a semantic category mask, its corresponding instance bounding box is predicted using a deep fully convolutional regression network. Thus it follows a different pipeline to the popular detect-then-segment approaches that first predict instances' bounding boxes, which are the current state-of-the-art in instance segmentation. We show that, by leveraging the strength of our state-of-the-art semantic segmentation models, the proposed method can achieve comparable or even better results to detect-then-segment approaches. We make the following contributions. (i) First, we propose a simple yet effective approach to semantic instance segmentation. (ii) Second, we propose an online bootstrapping method during training, which is critically important for achieving good performance for both semantic category segmentation and instance-level segmentation. (iii) As the performance of semantic category segmentation has a significant impact on the instance-level segmentation, which is the second step of our approach, we train fully convolutional residual networks to achieve the best semantic category segmentation accuracy. On the PASCAL VOC 2012 dataset, we obtain the currently best mean intersection-over-union score of 79.1%. (iv) We also achieve state-of-the-art results for instance-level segmentation.