Deep Watershed Transform for Instance Segmentation
arXiv:1611.08303
Abstract
Most contemporary approaches to instance segmentation use complex pipelines involving conditional random fields, recurrent neural networks, object proposals, or template matching schemes. In our paper, we present a simple yet powerful end-to-end convolutional neural network to tackle this task. Our approach combines intuitions from the classical watershed transform and modern deep learning to produce an energy map of the image where object instances are unambiguously represented as basins in the energy map. We then perform a cut at a single energy level to directly yield connected components corresponding to object instances. Our model more than doubles the performance of the state-of-the-art on the challenging Cityscapes Instance Level Segmentation task.
References in corpus (6)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- The Cityscapes Dataset for Semantic Urban Scene Understanding
- Fully Convolutional Networks for Semantic Segmentation
- Pyramid Scene Parsing Network
- Simultaneous Detection and Segmentation
- Pixel-level Encoding and Depth Layering for Instance-level Semantic Labeling
Cited by in corpus (18)
- Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics
- Semantic Instance Segmentation with a Discriminative Loss Function
- Path Aggregation Network for Instance Segmentation
- Evolution of Image Segmentation using Deep Convolutional Neural Network: A Survey
- Semantic Instance Segmentation via Deep Metric Learning
- Fast Scene Understanding for Autonomous Driving
- Learning deep structured active contours end-to-end
- Instance-level Human Parsing via Part Grouping Network
- Learning to Segment Every Thing
- PatchPerPix for Instance Segmentation
- Boundary-aware Instance Segmentation
- An Auxiliary Task for Learning Nuclei Segmentation in 3D Microscopy Images
- Instance Shadow Detection
- Primitive Fitting Using Deep Boundary Aware Geometric Segmentation
- Super-BPD: Super Boundary-to-Pixel Direction for Fast Image Segmentation
- DeepFlux for Skeletons in the Wild
- Pixelwise Instance Segmentation with a Dynamically Instantiated Network
- Transformer Assisted Convolutional Network for Cell Instance Segmentation