You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This repo is implementation for PointNet in pytorch https://arxiv.org/abs/1612.00593 in pytorch. The model is in file `pointnet.py`.
3
+
4
+
5
+
# Download data and running
6
+
7
+
```
8
+
bash build.sh #build C++ code for visualization
9
+
bash download.sh #download dataset
10
+
python train_classification.py #train 3D model classification
11
+
python python train_segmentation.py # train 3D model segmentaion
12
+
13
+
python show_seg.py --model seg/seg_model_20.pth # show segmentation results
14
+
```
15
+
16
+
# Performance
17
+
Without heavy tuning, PointNet can achieve 80-90% performance in classification and segmentaion on this [dataset](http://web.stanford.edu/~ericyi/project_page/part_annotation/index.html).
0 commit comments