Skip to content
Switch branches/tags

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time

Point Cloud Segmentation

By Soumik Rakshit & Sayak Paul


This repository provides a TF2 implementation of PointNet1 for segmenting point clouds. Our implementation is fully supported on TPUs allowing you to train models faster. Distributed training (single-device multi-worker) on GPUs is also supported and so is single-GPU training. For results and pre-trained models please see below.

To get an understanding of PointNet for segmentation, follow this blog post from Point cloud segmentation with PointNet.

We use the ShapeNetCore dataset to train our models on individual categories. The dataset is available here. To train and test our code, you don't need to download the dataset beforehand, though.

Update November 16, 2021: We won the #TFCommunitySpolight award for this project.

Running using Docker

  • Build image using docker build -t point-cloud-image .

  • Run Jupyter Server using docker run -it --gpus all -p 8888:8888 -v $(pwd):/usr/src/point-cloud-segmentation point-cloud-image

Create TFRecords for ShapeNetCore Shape Segmentation

This part is only required if you would like to train models using TPUs. Be advised that training using TPUs is usually recommended when you have sufficient amount of data. Therefore, you should only use TPUs for the following object categories:

  • Airplane
  • Car
  • Chair
  • Table

As such we only provide results and models for these categories.

Usage: [OPTIONS]

  python --experiment_configs configs/

Train for ShapeNetCore Shape Segmentation

Usage: [OPTIONS]

  --experiment_configs    Experiment configs (configs/
  --wandb_project_name    Project Name (DEFAULT: pointnet_shapenet_core)
  --use_wandb             Use WandB flag (DEFAULT: True)

  python --experiment_configs configs/

In case you want to change the configuration-related parameters, either edit them directly in configs/ or add a new configuration and specify the name of the configuration in the command line.

Notes on the Training Setup

  • The batch_size in the configuration denotes local batch size. If you are using single-host multi-worker distributed training, the batch_size denoted here will be multiplied by the number of workers you have.
  • Using a Google Cloud Storage (GCS) based artifact_location is not a requirement if you are using GPU(s). But for TPUs, it's a requirement.


We also provide notebooks for training and testing the models:

  • notebooks/train_gpu.ipynb lets you train using GPU(s). If you are using multiple GPUs in the single machine it will be detected automatically. If your machine supports mixed-precision, then also it will be detected automatically.
  • notebooks/train_tpu.ipynb lets you train using TPUs. For this using TFRecords for handling data IS a requirement.
  • notebooks/run_inference.ipynb lets you test the models on GPU(s) on individual object categories.
  • notebooks/keras-tuner.ipynb lets you tune the hyperparameters of the training routine namely number of epochs, initial learning rate (LR), and LR decaying epochs. We use Keras Tuner for this.

We track our training results using Weights and Biases (WandB). For the hyperparameter tuning part, we combine TensorBoard and WandB.

Segmentation Results and Models

Object Category

Training Result

Final Model

Airplane WandB Run SavedModel Link
Car WandB Run SavedModel Link
Chair WandB Run SavedModel Link
Table WandB Run SavedModel Link

Below are some segmentation results:






[1] PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation; Charles R. Qi, Hao Su, Kaichun Mo, Leonidas J. Guibas; CVPR 2017;


We are thankful to the GDE program for providing us GCP credits.