Semantic Instance Segmentation via Deep Metric Learning

TL;DR

This paper introduces a novel approach for semantic instance segmentation using deep metric learning to compute pixel similarity and group pixels, achieving competitive results on Pascal VOC.

Contribution

It presents a new deep metric learning-based method for semantic instance segmentation that combines pixel similarity scoring with seed-based grouping.

Findings

Achieved competitive results on Pascal VOC benchmark.

Demonstrated effectiveness of deep metric learning for pixel grouping.

Proposed a seed-based grouping method for instance segmentation.

Abstract

We propose a new method for semantic instance segmentation, by first computing how likely two pixels are to belong to the same object, and then by grouping similar pixels together. Our similarity metric is based on a deep, fully convolutional embedding model. Our grouping method is based on selecting all points that are sufficiently similar to a set of "seed points", chosen from a deep, fully convolutional scoring model. We show competitive results on the Pascal VOC instance segmentation benchmark.