Active Boundary Loss for Semantic Segmentation
Explore this paper's citation graph
Summary
Experimental results show that training with the active boundary loss can effectively improve the boundary F-score and mean Intersection-over-Union on challenging image and video object segmentation datasets.
- Type
- article
- Published
- 2021-02-04
- Cited by
- 103
- References
- 78
- Access
- Open access
- OpenAlex
- https://openalex.org/W3127595489
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:231802302
Keywords
Boundary (topology), Segmentation, Ground truth, Computer science, Artificial intelligence
References
- Snakes, shapes, and gradient vector flow
- Fully convolutional networks for semantic segmentation
- The Pascal Visual Object Classes Challenge: A Retrospective
- Snakes: Active contour models
- Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data
- Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials
- Rethinking the Inception Architecture for Computer Vision
- Deep Residual Learning for Image Recognition
- Semantic Segmentation with Boundary Neural Fields
- The Cityscapes Dataset for Semantic Urban Scene Understanding
- DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs
- A Benchmark Dataset and Evaluation Methodology for Video Object Segmentation
- Roto++
- One-Shot Video Object Segmentation
- Fully Convolutional Instance-Aware Semantic Segmentation
- Pyramid Scene Parsing Network
- Video Propagation Networks
- RefineNet: Multi-path Refinement Networks for High-Resolution Semantic Segmentation
- Large Kernel Matters — Improve Semantic Segmentation by Global Convolutional Network
- Rethinking Atrous Convolution for Semantic Image Segmentation
Cited by
- Standardized Max Logits: A Simple yet Effective Approach for Identifying Unexpected Road Obstacles in Urban-Scene Segmentation
- Region-wise Loss for Biomedical Image Segmentation
- Zero Pixel Directional Boundary by Vector Transform
- Introducing the Boundary-Aware loss for deep image segmentation
- 3D Convolutional Neural Networks for Dendrite Segmentation Using Fine-Tuning and Hyperparameter Optimization
- Wave Loss: A Topographic Metric for Image Segmentation
- A Stronger Baseline for Seismic Facies Classification With Less Data
- Boosting semantic segmentation via feature enhancement
- Improving Image Segmentation with Boundary Patch Refinement
- Context and Boundary Guided Multi-Scale Feature Fusion Network for Semantic Segmentation
- Contour-Aware Equipotential Learning for Semantic Segmentation
- Coarse-to-Fine Feature Mining for Video Semantic Segmentation
- Representation Separation for Semantic Segmentation with Vision Transformers
- Semantic Diffusion Network for Semantic Segmentation
- Epistemic and aleatoric uncertainty quantification for crack detection using a Bayesian Boundary Aware Convolutional Network
- PCCN: Progressive Context Comprehension Network for Oil Stains Detection of High-Speed Trains
- Improving Semantic Segmentation via Decoupled Body and Edge Information
- Unsupervised domain adaptation via style adaptation and boundary enhancement for medical semantic segmentation
- Conditional Boundary Loss for Semantic Segmentation
- TriangleNet: Edge Prior Augmented Network for Semantic Segmentation through Cross-Task Consistency
Related papers
- Object Detection, Tracking, and Motion Segmentation for Object-level Video Segmentation
- Segmentation of moving images by the human visual system
- Object segmentation using stereo images
- Physical interaction for segmentation of unknown textured and non-textured rigid objects
- Wavelet-based skin segmentation for detecting occluded arms in human body pose modelling system
- Contour extraction of moving objects
- Evaluation and Improvement of Region-Based Motion Segmentation
- Entropy Based ROI Extraction and Modified Contour Model for Image Segmentation
- Video segmentation using active contours on a group of pictures
- Left ventricular segmentation of cardiac magnetic resonance images using a novel edge following technique