End-to-end detection-segmentation network with ROI convolution

TL;DR

This paper introduces an end-to-end detection-segmentation network that enhances segmentation accuracy by integrating object localization as a guiding cue, demonstrated on ultrasound images of small objects.

Contribution

It presents a novel joint detection and segmentation network with a localization unit, improving segmentation performance over traditional methods.

Findings

Improved segmentation accuracy on ultrasound images

Joint learning of detection and segmentation enhances results

Code is publicly available for reproducibility

Abstract

We propose an end-to-end neural network that improves the segmentation accuracy of fully convolutional networks by incorporating a localization unit. This network performs object localization first, which is then used as a cue to guide the training of the segmentation network. We test the proposed method on a segmentation task of small objects on a clinical dataset of ultrasound images. We show that by jointly learning for detection and segmentation, the proposed network is able to improve the segmentation accuracy compared to only learning for segmentation. Code is publicly available at https://github.com/vincentzhang/roi-fcn.