All Projects → driving-behavior → Dbnet

driving-behavior / Dbnet

Licence: apache-2.0
DBNet: A Large-Scale Dataset for Driving Behavior Learning, CVPR 2018

Programming Languages

python
139335 projects - #7 most used programming language

Projects that are alternatives of or similar to Dbnet

LiDAR fog sim
LiDAR fog simulation
Stars: ✭ 101 (-41.28%)
Mutual labels:  point-cloud, autonomous-driving
awesome-lidar
😎 Awesome LIDAR list. The list includes LIDAR manufacturers, datasets, point cloud-processing algorithms, point cloud frameworks and simulators.
Stars: ✭ 217 (+26.16%)
Mutual labels:  point-cloud, autonomous-driving
FLAT
[ICCV2021 Oral] Fooling LiDAR by Attacking GPS Trajectory
Stars: ✭ 52 (-69.77%)
Mutual labels:  point-cloud, autonomous-driving
3d Bat
3D Bounding Box Annotation Tool (3D-BAT) Point cloud and Image Labeling
Stars: ✭ 179 (+4.07%)
Mutual labels:  autonomous-driving, point-cloud
Pointcnn
PointCNN: Convolution On X-Transformed Points (NeurIPS 2018)
Stars: ✭ 1,120 (+551.16%)
Mutual labels:  autonomous-driving, point-cloud
Depth-Guided-Inpainting
Code for ECCV 2020 "DVI: Depth Guided Video Inpainting for Autonomous Driving"
Stars: ✭ 50 (-70.93%)
Mutual labels:  point-cloud, autonomous-driving
urban road filter
Real-time LIDAR-based Urban Road and Sidewalk detection for Autonomous Vehicles 🚗
Stars: ✭ 134 (-22.09%)
Mutual labels:  point-cloud, autonomous-driving
3d Pointcloud
Papers and Datasets about Point Cloud.
Stars: ✭ 179 (+4.07%)
Mutual labels:  autonomous-driving, point-cloud
So Net
SO-Net: Self-Organizing Network for Point Cloud Analysis, CVPR2018
Stars: ✭ 297 (+72.67%)
Mutual labels:  autonomous-driving, point-cloud
Splatnet
SPLATNet: Sparse Lattice Networks for Point Cloud Processing (CVPR2018)
Stars: ✭ 259 (+50.58%)
Mutual labels:  point-cloud, cvpr2018
efficient online learning
Efficient Online Transfer Learning for 3D Object Detection in Autonomous Driving
Stars: ✭ 20 (-88.37%)
Mutual labels:  point-cloud, autonomous-driving
Awesome Robotic Tooling
Tooling for professional robotic development in C++ and Python with a touch of ROS, autonomous driving and aerospace.
Stars: ✭ 1,876 (+990.7%)
Mutual labels:  autonomous-driving, point-cloud
BtcDet
Behind the Curtain: Learning Occluded Shapes for 3D Object Detection
Stars: ✭ 104 (-39.53%)
Mutual labels:  point-cloud, autonomous-driving
Dh3d
DH3D: Deep Hierarchical 3D Descriptors for Robust Large-Scale 6DOF Relocalization
Stars: ✭ 96 (-44.19%)
Mutual labels:  autonomous-driving, point-cloud
Openpcdet
OpenPCDet Toolbox for LiDAR-based 3D Object Detection.
Stars: ✭ 2,199 (+1178.49%)
Mutual labels:  autonomous-driving, point-cloud
Kubestone
Performance benchmarks for Kubernetes
Stars: ✭ 159 (-7.56%)
Mutual labels:  benchmark
Gym Carla
An OpenAI gym wrapper for CARLA simulator
Stars: ✭ 164 (-4.65%)
Mutual labels:  autonomous-driving
Pangolin
Python binding of 3D visualization library Pangolin
Stars: ✭ 157 (-8.72%)
Mutual labels:  point-cloud
Blue benchmark
BLUE benchmark consists of five different biomedicine text-mining tasks with ten corpora.
Stars: ✭ 159 (-7.56%)
Mutual labels:  benchmark
Dlbench
Benchmarking State-of-the-Art Deep Learning Software Tools
Stars: ✭ 169 (-1.74%)
Mutual labels:  benchmark

db-prediction

DBNet is a large-scale driving behavior dataset, which provides large-scale high-quality point clouds scanned by Velodyne lasers, high-resolution videos recorded by dashboard cameras and standard drivers' behaviors (vehicle speed, steering angle) collected by real-time sensors.

Extensive experiments demonstrate that extra depth information helps networks to determine driving policies indeed. We hope it will become useful resources for the autonomous driving research community.

Created by Yiping Chen*, Jingkang Wang*, Jonathan Li, Cewu Lu, Zhipeng Luo, HanXue and Cheng Wang. (*equal contribution)

The resources of our work are available: [paper], [code], [video], [website], [challenge], [prepared data]

Contents

  1. Introduction
  2. Requirements
  3. Quick Start
  4. Baseline
  5. Contributors
  6. Citation
  7. License

Introduction

This work is based on our research paper, which appears in CVPR 2018. We propose a large-scale dataset for driving behavior learning, namely, DBNet. You can also check our dataset webpage for a deeper introduction.

In this repository, we release demo code and partial prepared data for training with only images, as well as leveraging feature maps or point clouds. The prepared data are accessible here. (More demo models and scripts are released soon!)

Requirements

  • Tensorflow 1.2.0
  • Python 2.7
  • CUDA 8.0+ (For GPU)
  • Python Libraries: numpy, scipy and laspy

The code has been tested with Python 2.7, Tensorflow 1.2.0, CUDA 8.0 and cuDNN 5.1 on Ubuntu 14.04. But it may work on more machines (directly or through mini-modification), pull-requests or test report are well welcomed.

Quick Start

Training

To train a model to predict vehicle speeds and steering angles:

python train.py --model nvidia_pn --batch_size 16 --max_epoch 125 --gpu 0

The names of the models are consistent with our paper. Log files and network parameters will be saved to logs folder in default.

To see HELP for the training script:

python train.py -h

We can use TensorBoard to view the network architecture and monitor the training progress.

tensorboard --logdir logs

Evaluation

After training, you could evaluate the performance of models using evaluate.py. To plot the figures or calculate AUC, you may need to have matplotlib library installed.

python evaluate.py --model_path logs/nvidia_pn/model.ckpt

Prediction

To get the predictions of test data:

python predict.py

The results are saved in results/results (every segment) and results/behavior_pred.txt (merged) by default. To change the storation location:

python predict.py --result_dir specified_dir

The result directory will be created automatically if it doesn't exist.

Baseline

Method Setting Accuracy AUC ME AE AME
nvidia-pn Videos + Laser Points angle 70.65% (<5) 0.7799 29.46 4.23 20.88
speed 82.21% (<3) 0.8701 18.56 1.80 9.68

This baseline is run on dbnet-2018 challenge data and only nvidia_pn is tested. To measure difficult architectures comprehensively, several metrics are set, including accuracy under different thresholds, area under curve (AUC), max error (ME), mean error (AE) and mean of max errors (AME).

The implementations of these metrics could be found in evaluate.py.

Contributors

DBNet was developed by MVIG, Shanghai Jiao Tong University* and SCSC Lab, Xiamen University* (alphabetical order).

Citation

If you find our work useful in your research, please consider citing:

@InProceedings{DBNet2018,
  author = {Yiping Chen and Jingkang Wang and Jonathan Li and Cewu Lu and Zhipeng Luo and HanXue and Cheng Wang},
  title = {LiDAR-Video Driving Dataset: Learning Driving Policies Effectively},
  booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
  month = {June},
  year = {2018}
}

License

Our code is released under Apache 2.0 License. The copyright of DBNet could be checked here.

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].