PolyTransform: Deep Polygon Transformer for Instance Segmentation
arXiv:1912.02801
Abstract
In this paper, we propose PolyTransform, a novel instance segmentation algorithm that produces precise, geometry-preserving masks by combining the strengths of prevailing segmentation approaches and modern polygon-based methods. In particular, we first exploit a segmentation network to generate instance masks. We then convert the masks into a set of polygons that are then fed to a deforming network that transforms the polygons such that they better fit the object boundaries. Our experiments on the challenging Cityscapes dataset show that our PolyTransform significantly improves the performance of the backbone instance segmentation network and ranks 1st on the Cityscapes test-set leaderboard. We also show impressive gains in the interactive annotation setting. We release the code at https://github.com/uber-research/PolyTransform.
References in corpus (22)
- Adam: A Method for Stochastic Optimization
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
- Rethinking Atrous Convolution for Semantic Image Segmentation
- Deep Residual Learning for Image Recognition
- Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials
- Learning to Segment Object Candidates
- Multiscale Combinatorial Grouping for Image Segmentation and Object Proposal Generation
- Associative Embedding: End-to-End Learning for Joint Detection and Grouping
- Horovod: fast and easy distributed deep learning in TensorFlow
- Convolutional Feature Masking for Joint Object and Stuff Segmentation
- Feature Pyramid Networks for Object Detection
- Deformable Convolutional Networks
- Path Aggregation Network for Instance Segmentation
- Hybrid Task Cascade for Instance Segmentation
- Annotating Object Instances with a Polygon-RNN
- SSAP: Single-Shot Instance Segmentation With Affinity Pyramid
- Fast Interactive Object Annotation with Curve-GCN
- ShapeMask: Learning to Segment Novel Objects by Refining Shape Priors
- End-to-End Deep Structured Models for Drawing Crosswalks
- Deep Extreme Cut: From Extreme Points to Object Segmentation
- DARNet: Deep Active Ray Network for Building Segmentation
- Deep Rigid Instance Scene Flow