chengdazhi/Deformable-Convolution-V2-PyTorch
PyTorch implementation of deformable convolution operations for computer vision tasks.
Not currently ranked — collecting fresh signals.
star history
This repository provides a PyTorch and CUDA implementation of Deformable Convolution V2, a computer vision technique that enables the convolution kernel to sample from arbitrary locations with learned offsets. It is ported from the original MXNet implementation and is used as a component in object detection frameworks like mmdetection for tasks such as object detection and segmentation.
Frequently asked
- What is chengdazhi/Deformable-Convolution-V2-PyTorch?
- PyTorch implementation of deformable convolution operations for computer vision tasks.
- Is Deformable-Convolution-V2-PyTorch open source?
- Yes — chengdazhi/Deformable-Convolution-V2-PyTorch is open source, released under the MIT license.
- What language is Deformable-Convolution-V2-PyTorch written in?
- chengdazhi/Deformable-Convolution-V2-PyTorch is primarily written in Cuda.
- How popular is Deformable-Convolution-V2-PyTorch?
- chengdazhi/Deformable-Convolution-V2-PyTorch has 1.5k stars on GitHub.
- Where can I find Deformable-Convolution-V2-PyTorch?
- chengdazhi/Deformable-Convolution-V2-PyTorch is on GitHub at https://github.com/chengdazhi/Deformable-Convolution-V2-PyTorch.