Abstract:Although great success has been achieved in instance segmentation, accurate segmentation of instances remains difficult, especially at object edges. This problem is more prominent for instance segmentation in remote sensing imagery due to the diverse scales, variable illumination, smaller objects, and complex backgrounds. We find that most current instance segmentation networks do not consider the segmentation difficulty of different instances and different regions within the instance. In this paper, we study this problem and propose an ensemble method to segment instances from remote sensing images, considering the enhancement of hard-to-segment instances and instance edges. First, we apply a pixel-level Dice metric that reliably describes the segmentation quality of each instance to achieve online hard instance learning. Instances with low Dice values are studied with emphasis. Second, we generate a penalty map based on the analysis of boundary shapes to not only enhance the edges of objects but also discriminatively strengthen the edges of different shapes. That is, different areas of an object, such as internal areas, flat edges, and sharp edges, are distinguished and discriminatively weighed. Finally, the hard-to-segment instance learning and the shape-penalty map are integrated for precise instance segmentation. To evaluate the effectiveness and generalization ability of the proposed method, we train with the classic instance segmentation network Mask R-CNN and conduct experiments on two different types of remote sensing datasets: the iSAID-Reduce100 and the JKGW_WHU datasets, which have extremely different feature distributions and spatial resolutions. The comprehensive experimental results show that the proposed method improved the segmentation results by 2.78% and 1.77% in mask AP on the iSAID-Reduce100 and JKGW_WHU datasets, respectively. We also test other state-of-the-art (SOTA) methods that focus on inaccurate edges. Experiments demonstrate that our method outperforms these methods.

Instance segmentation method based on target contour

Instance segmentation method based on target contour

A Simultaneous Object Detection and Component Segmentation Approach Based on Mask R-CNN

Instance-Level Salient Object Segmentation.

Contour Loss for Instance Segmentation via k-step Distance Transformation Image

Associate Semantic-Instance Segmentation of 3D Point Clouds Based on Local Feature Extraction

Instance Segmentation in Very High Resolution Remote Sensing Imagery Based on Hard-to-Segment Instance Learning and Boundary Shape Analysis

Segmenting Beyond the Bounding Box for Instance Segmentation.

A Depth Information Aided Real-Time Instance Segmentation Method for Space Task Scenarios under CPU Platform

Contour-Assisted Long-Range Perceptual Network for Camouflaged Instance Segmentation

Instance Mask Growing on Leaf.

Joint object contour points and semantics for instance segmentation

Contour and Enclosed Region Refining for Contour-Based Instance Segmentation

A Two-Pipeline Instance Segmentation Network via Boundary Enhancement for Scene Understanding

Growing Instance Mask on Leaf

A Super-pixel Based Method for Instance Segmentation Post-processing

Object-Guided Instance Segmentation With Auxiliary Feature Refinement for Biological Images

Supervised Edge Attention Network for Accurate Image Instance Segmentation

Object-Guided Instance Segmentation for Biological Images

Real-Time Instance Segmentation Method Based on Location Attention