NVlabs/PWC-Net
PWC-Net is a CNN-based optical flow estimation model using pyramid, warping, and cost volume techniques.
Not currently ranked — collecting fresh signals.
star history
PWC-Net is an end-to-end trainable deep neural network for optical flow estimation, published at CVPR 2018. It fuses classic computer vision techniques—image pyramid, warping, and cost volume—into a single model to achieve state-of-the-art results. The repository provides implementations in both Caffe and PyTorch, nearly matching in accuracy.
Frequently asked
- What is NVlabs/PWC-Net?
- PWC-Net is a CNN-based optical flow estimation model using pyramid, warping, and cost volume techniques.
- Is PWC-Net open source?
- Yes — NVlabs/PWC-Net is an open-source project tracked on heatdrop.
- What language is PWC-Net written in?
- NVlabs/PWC-Net is primarily written in Python.
- How popular is PWC-Net?
- NVlabs/PWC-Net has 1.7k stars on GitHub.
- Where can I find PWC-Net?
- NVlabs/PWC-Net is on GitHub at https://github.com/NVlabs/PWC-Net.