art-programmer/PlaneNet
A neural network for piece-wise planar 3D reconstruction from a single RGB image.
Not currently ranked — collecting fresh signals.
star history
PlaneNet is an end-to-end deep learning architecture that infers plane parameters and corresponding segmentation masks from a single RGB image. The model performs geometric scene understanding by detecting and segmenting planar surfaces in 3D space. It was published at CVPR 2018 and includes both TensorFlow and experimental PyTorch implementations for training and testing.
Frequently asked
- What is art-programmer/PlaneNet?
- A neural network for piece-wise planar 3D reconstruction from a single RGB image.
- Is PlaneNet open source?
- Yes — art-programmer/PlaneNet is open source, released under the MIT license.
- What language is PlaneNet written in?
- art-programmer/PlaneNet is primarily written in Python.
- How popular is PlaneNet?
- art-programmer/PlaneNet has 422 stars on GitHub.
- Where can I find PlaneNet?
- art-programmer/PlaneNet is on GitHub at https://github.com/art-programmer/PlaneNet.