https://github.com/yanx27/3DGNN_pytorch

3D Graph Neural Networks for RGBD Semantic Segmentation
https://github.com/yanx27/3DGNN_pytorch

gnns point-cloud pytorch rgbd segmentation

Last synced: 7 days ago
JSON representation

3D Graph Neural Networks for RGBD Semantic Segmentation

• Host: GitHub
• URL: https://github.com/yanx27/3DGNN_pytorch
• Owner: yanx27
• License: mit
• Created: 2018-11-25T15:08:29.000Z (almost 6 years ago)
• Default Branch: master
• Last Pushed: 2019-03-16T12:55:46.000Z (over 5 years ago)
• Last Synced: 2024-08-01T03:46:10.934Z (3 months ago)
• Topics: gnns, point-cloud, pytorch, rgbd, segmentation
• Language: Python
• Homepage:
• Size: 193 KB
• Stars: 229
• Watchers: 5
• Forks: 43
• Open Issues: 15
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# 3DGNN for RGB-D segmentation

This is the Pytorch implementation of [3D Graph Neural Networks for RGBD Semantic Segmentation](http://openaccess.thecvf.com/content_ICCV_2017/papers/Qi_3D_Graph_Neural_ICCV_2017_paper.pdf): 

### Data Preparation

1. Download NYU_Depth_V2 dataset from [here](https://cs.nyu.edu/~silberman/datasets/nyu_depth_v2.html) and select scenes and save as `./datasets/data/nyu_depth_v2_labeled.mat`

2. Transfer depth images to hha by yourself from [here](https://github.com/charlesCXK/Depth2HHA) and save in `./datasets/data/hha/`. 

### Emviroment

Required CUDA (8.0) + pytorch 0.4.1