Abstract:An effective semantic segmentation approach should contain fine object boundaries and continuous regions. However, recent mask-based segmentation cannot extract boundary features well on the coarse prediction, which causes obvious problems of blurry edges. Although several segmentation methods embed boundary de-tection branch to calculate contours directly, this type of architecture will lead to the increased computational complexity and miss the edge detailed information such as inter-class distinction. In order to obtain fine boundaries, we present a lightweight boundary refinement module with point supervision named BRPS to im-prove the boundary quality for the segmentation result generated by various existing segmentation models. Firstly, a direction field is learned to complete the initial feature rectification, which is defined as pointing away from the nearest object boundary to each pixel, where the weighted Euclidean and Cosine distance func-tion is used as the loss between predicted boundary pixels and Ground-truth labels. Then, point-based super-vised learning is performed at uncertain and certain locations including random distribution and key feature points based on a new point convolutional operation to output final crisp object boundaries. Finally, we verify that our BRPS module can effectively reduce the prediction errors for segmentation results generated from various state-of-the-art models such as DeepLabv3 and HRNet on the Pascal VOC2012, NYUD v2 datasets, City-scapes and BDD100K datasets.& nbsp; (c) 2021 Elsevier B.V. All rights reserved.

Boundary Perception Guidance: A Scribble-Supervised Semantic Segmentation Approach.

Scribble-Supervised Segmentation of Aerial Building Footprints Using Adversarial Learning

Scribble-Supervised Semantic Segmentation Inference

Salient-Boundary-Guided Pseudo-Pixel Supervision for Weakly-Supervised Semantic Segmentation

Improving Image Segmentation with Boundary Patch Refinement

Boundary-aware Graph Convolution for Semantic Segmentation.

Scribble-Based Medical Image Segmentation Via Shape Perturbation Consistency and Boundary Enhancement Constraint

Improving Semantic Image Segmentation with a Probabilistic Superpixel-Based Dense Conditional Random Field.

Weakly Supervised Semantic Segmentation with Boundary Exploration

BASeg: Boundary aware semantic segmentation for autonomous driving

Semantic Segmentation Via Structured Patch Prediction, Context Crf And Guidance Crf

Boundary-Aware Geometric Encoding for Semantic Segmentation of Point Clouds

Lightweight boundary refinement module based on point supervision for semantic segmentation

BLPSeg: Balance the Label Preference in Scribble-Supervised Semantic Segmentation

Bounding Box Based Annotation Generation For Semantic Segmentation By Boundary Detection

Cross-Pixel Dependency with Boundary-Feature Transformation for Weakly Supervised Semantic Segmentation

Scribble Hides Class: Promoting Scribble-Based Weakly-Supervised Semantic Segmentation with Its Class Label

Push-the-Boundary: Boundary-aware Feature Propagation for Semantic Segmentation of 3D Point Clouds

ZScribbleSeg: Zen and the Art of Scribble Supervised Medical Image Segmentation

Semantic Segmentation Via Structured Refined Prediction and Dual Global Priors.

ScribbleVS: Scribble-Supervised Medical Image Segmentation via Dynamic Competitive Pseudo Label Selection