Deep Watershed Transform for Instance Segmentation

TL;DR

This paper introduces a simple end-to-end CNN that combines classical watershed concepts with deep learning to improve instance segmentation, significantly outperforming previous methods on Cityscapes.

Contribution

It presents a novel approach that unifies watershed transform principles with deep learning for efficient instance segmentation.

Findings

More than doubled the performance on Cityscapes.

Produced clear object instance basins in energy maps.

Achieved state-of-the-art results with a simple pipeline.

Abstract

Most contemporary approaches to instance segmentation use complex pipelines involving conditional random fields, recurrent neural networks, object proposals, or template matching schemes. In our paper, we present a simple yet powerful end-to-end convolutional neural network to tackle this task. Our approach combines intuitions from the classical watershed transform and modern deep learning to produce an energy map of the image where object instances are unambiguously represented as basins in the energy map. We then perform a cut at a single energy level to directly yield connected components corresponding to object instances. Our model more than doubles the performance of the state-of-the-art on the challenging Cityscapes Instance Level Segmentation task.