Vec2Instance: Parameterization for Deep Instance Segmentation

TL;DR

Vec2Instance introduces a novel deep learning architecture for instance segmentation that parametrizes instances around their centroids, achieving high accuracy on satellite imagery with a simpler approach than existing methods.

Contribution

The paper presents Vec2Instance, a new neural network architecture that efficiently estimates complex instance shapes using parametrization, offering an alternative to complex segmentation pipelines.

Findings

Achieved 89% pixel-wise accuracy on satellite images.

Demonstrated effectiveness in extracting building footprints.

Comparable performance to Mask R-CNN with simpler architecture.

Abstract

Current advances in deep learning is leading to human-level accuracy in computer vision tasks such as object classification, localization, semantic segmentation, and instance segmentation. In this paper, we describe a new deep convolutional neural network architecture called Vec2Instance for instance segmentation. Vec2Instance provides a framework for parametrization of instances, allowing convolutional neural networks to efficiently estimate the complex shapes of instances around their centroids. We demonstrate the feasibility of the proposed architecture with respect to instance segmentation tasks on satellite images, which have a wide range of applications. Moreover, we demonstrate the usefulness of the new method for extracting building foot-prints from satellite images. Total pixel-wise accuracy of our approach is 89\%, near the accuracy of the state-of-the-art Mask RCNN (91\%). Vec2Instance is an alternative approach to complex instance segmentation pipelines, offering simplicity and intuitiveness. The code developed under this study is available in the Vec2Instance GitHub repository, https://github.com/lakmalnd/Vec2Instance