CaDDN

This is the unofficial implementation of the paper CaDDN for personal study. This codebase is based on https://github.com/xinzhuma/monodle.

Usage

Installation

This repo is tested on our local environment (python=3.7, cuda=10.1, pytorch=1.4)

conda create -n CaDDN python=3.7

Then, activate the environment:

conda activate CaDDN

Install PyTorch:

conda install pytorch==1.4.0 torchvision==0.5.0 cudatoolkit=10.1 -c pytorch

and other requirements:

pip install -r requirements.txt

python setup.py develop

Data Preparation

Please download KITTI dataset and organize the data as follows:

#ROOT
  |data/
    |KITTI/
      |ImageSets/
      |object/			
        |training/
          |calib/
          |image_2/
          |velodyne/
          |label/
        |testing/
          |calib/
          |image_2/
          |image_3/

Training

Move to the workplace and train the network:

 cd #ROOT
 cd experiments/example
 CUDA_VISIBLE_DEVICES=0,1 bash ./train.sh 2 kitti_example.yaml