Spatio-Temporal Segmentation

This repository contains the accompanying code for 4D-SpatioTemporal ConvNets: Minkowski Convolutional Neural Networks, CVPR'19.

Change Log



You need to install pytorch and Minkowski Engine either with pip or with anaconda.


The MinkowskiEngine is distributed via PyPI MinkowskiEngine which can be installed simply with pip. First, install pytorch following the instruction. Next, install openblas.

sudo apt install libopenblas-dev

pip install torch torchvision

pip install -U git+

Next, clone the repository and install the rest of the requirements

git clone

cd SpatioTemporalSegmentation

pip install -r requirements.txt


Please visit the MinkowskiEngine issue pages if you have difficulties installing Minkowski Engine.

ScanNet Training

  1. Download the ScanNet dataset from the official website. You need to sign the terms of use.

  2. Next, preprocess all scannet raw point cloud with the following command after you set the path correctly.

python -m lib.datasets.preprocessing.scannet
  1. Train the network with
export BATCH_SIZE=N;
./scripts/ 0 \
    -default \
    "--scannet_path /path/to/preprocessed/scannet"

Modify the BATCH_SIZE accordingly.

The first argument is the GPU id and the second argument is the path postfix and the last argument is the miscellaneous arguments.

mIoU vs. Overall Accuracy

The official evaluation metric for ScanNet is mIoU. OA, Overal Accuracy is not the official metric since it is not discriminative. This is the convention from the 2D semantic segmentation as the pixelwise overall accuracy does not capture the fidelity of the semantic segmentation. On 3D ScanNet semantic segmentation, OA: 89.087 -> mIOU 71.496 mAP 76.127 mAcc 79.660 on the ScanNet validation set v2.

Then why is the overall accuracy least discriminative metric? This is due to the fact that most of the scenes consist of large structures such as walls, floors, or background and scores on these will dominate the statistics if you use Overall Accuracy.

Synthia 4D Experiment

  1. Download the dataset from download

  2. Extract

cd /path/to/extract/synthia4d
tar -xf Synthia4D.tar
tar -xvjf *.tar.bz2
  1. Training
export BATCH_SIZE=N; \
./scripts/ 0 \
    "-default" \
    "--synthia_path /path/to/extract/synthia4d"

The above script trains a network. You have to change the arguments accordingly. The first argument to the script is the GPU id. Second argument is the log directory postfix; change to mark your experimental setup. The final argument is a series of the miscellaneous aruments. You have to specify the synthia directory here. Also, you have to wrap all arguments with " ".

Stanford 3D Dataset

  1. Download the stanford 3d dataset from the website

  2. Preprocess

Modify the input and output directory accordingly in


And run

python -m lib.datasets.preprocessing.stanford
  1. Train
./scripts/ 0 \
    "-default" \
    "--stanford3d_path /PATH/TO/PREPROCESSED/STANFORD"

Model Zoo

Model Dataset Voxel Size Conv1 Kernel Size Performance Link
Mink16UNet34C ScanNet train + val 2cm 3 Test set 73.6% mIoU, no sliding window download
Mink16UNet34C ScanNet train 2cm 5 Val 72.219% mIoU, no rotation average, no sliding window per class performance download
Mink16UNet18 Stanford Area5 train 5cm 5 Area 5 test 65.828% mIoU, no rotation average, no sliding window per class performance download
Mink16UNet34 Stanford Area5 train 5cm 5 Area 5 test 66.348% mIoU, no rotation average, no sliding window per class performance download
3D Mink16UNet14A Synthia CVPR19 train 15cm 3 CVPR19 test 81.903% mIoU, no rotation average, no sliding window per class performance download
3D Mink16UNet18 Synthia CVPR19 train 15cm 3 CVPR19 test 82.762% mIoU, no rotation average, no sliding window per class performance download

Note that sliding window style evaluation (cropping and stitching results) used in many related works effectively works as an ensemble (rotation averaging) which boosts the performance.


The demo code will download weights and an example scene first and then visualize prediction results.

Dataset Scannet Stanford
Command python -m demo.scannet python -m demo.stanford

Citing this work

If you use the Minkowski Engine, please cite:

  title={4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks},
  author={Choy, Christopher and Gwak, JunYoung and Savarese, Silvio},
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},

Related projects