PointNet.pytorch

This repo is implementation for PointNet(https://arxiv.org/abs/1612.00593) in pytorch. The model is in pointnet/model.py.

It is tested with pytorch-1.0.

Download data and running

git clone https://github.com/fxia22/pointnet.pytorch
cd pointnet.pytorch
pip install -e .

Download and build visualization tool

cd script
bash build.sh #build C++ code for visualization
bash download.sh #download dataset

Training

cd utils
python train_classification.py --dataset <dataset path> --nepoch=<number epochs> --dataset_type <modelnet40 | shapenet>
python train_segmentation.py --dataset <dataset path> --nepoch=<number epochs>

Use --feature_transform to use feature transform.

Performance

Classification performance

On ModelNet40:

	Overall Acc
Original implementation	89.2
this implementation(w/o feature transform)	86.4
this implementation(w/ feature transform)	87.0

On A subset of shapenet

	Overall Acc
Original implementation	N/A
this implementation(w/o feature transform)	98.1
this implementation(w/ feature transform)	97.7

Segmentation performance

Segmentation on A subset of shapenet.

Class(mIOU)	Airplane	Bag	Cap	Car	Chair	Earphone	Guitar	Knife	Lamp	Laptop	Motorbike	Mug	Pistol	Rocket	Skateboard	Table
Original implementation	83.4	78.7	82.5	74.9	89.6	73.0	91.5	85.9	80.8	95.3	65.2	93.0	81.2	57.9	72.8	80.6
this implementation(w/o feature transform)	73.5	71.3	64.3	61.1	87.2	69.5	86.1	81.6	77.4	92.7	41.3	86.5	78.2	41.2	61.0	81.1
this implementation(w/ feature transform)					87.6											81.0

Note that this implementation trains each class separately, so classes with fewer data will have slightly lower performance than reference implementation.

Sample segmentation result: