Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
tjulitianyi 757ff90674 | 1 year ago | |
---|---|---|
.idea | 1 year ago | |
figure | 1 year ago | |
models | 1 year ago | |
part_seg | 1 year ago | |
scannet | 1 year ago | |
tf_ops | 1 year ago | |
utils | 1 year ago | |
.gitignore | 1 year ago | |
README.md | 1 year ago | |
evaluate.py | 1 year ago | |
evaluate_multigpu.py | 1 year ago | |
modelnet_dataset.py | 1 year ago | |
modelnet_h5_dataset.py | 1 year ago | |
train.py | 1 year ago | |
train_multi_gpu.py | 1 year ago |
The official implementation of LSANet in tensorflow.
Thanks to Xuan-Yi Li,and Deng-Ping Fan's help.
We propose a new network layer, named Local Spatial Aware (LSA) Layer, to model geometric structure
in local region accurately and robustly. Each feature extracting operation in LSA layer is related to Spatial Distribution Weights (SDWs), which are learned based on the spatial distribution in local region, to establish a strong link with inherent geometric shape. The experiments show that our LSA-based network, named LSANet, can achieve on par or better performance than the state-of-the-art methods when evaluating on the challenging benchmark datasets. The network architecture of LSANet and LSA module are shown below.
The code is based on PointNet++. Please follow the instruction in PointNet++ to compile the customized TF operators.
ModelNet40 dataset can be downloaded here.
To train a LSANet for classification, please run
python train_multi_gpu.py
ShapeNet dataset can be downloaded here. To train a LSANet for segmentation, please run
cd part_seg
python train_multi_gpu_one_hot.py
Please cite this paper if it is helpful to your research,
@article{chen2019lsanet,
title={LSANet: Feature Learning on Point Sets by Local Spatial Aware Layer},
author={Chen, Lin-Zhuo and Li, Xuan-Yi and Fan, Deng-Ping and Cheng, Ming-Ming and Wang, Kai and Lu, Shao-Ping},
journal={arXiv preprint arXiv:1905.05442},
year={2019}
}
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》