ADAS is short for Advanced Driver Assistance System. And it has the following features night vision assistance, anti-collision alert, collision resolution, lane departure warning, lane keeping, lane change assist, corner obstacle detection, driver status monitoring, driver reminder, parking assistance, traffic sign recognition, and high beam Auxiliary and so on.

From the system function, the ADAS system has a significant number of tasks related to computer vision (CV) which are as follows:

1. Detection: vehicle detection, pedestrian detection, non-motor vehicle detection, traffic sign recognition;
2. Segmentation: lane detection.

In order to realize the above functions, it is necessary to separately establish a model for each function. However, parameters for these models are so many that the performance of the vehicle real-time system may be influenced.

Python version: 3.5
Deep learning framework: pytorch

package list:

1. scikit-image (0.12.3)
2. numpy (1.16.2)
3. torch (1.0.1)
4. torchvision (0.2.2.post3)
5. visdom (0.1.8.8)
6. matplotlib (1.5.3)
7. pandas (0.18.1)
8. opencv-python (3.4.3.18)
9. tqdm (4.26.0)

Lane Segmentation: https://xingangpan.github.io/projects/CULane.html

• Segmentation
  • Traditional segmentation method
  • 2D U-Net Seg
  • VPGNet
  • S-CNN
  • Least Square Fitting
• Detection
  • YOLO
  • Faster-RCNN
  • Stereo-RCNN
• Combining the above

1. Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation[C]//International Conference on Medical image computing and computer-assisted intervention. Springer, Cham, 2015: 234-241.
2. Pan X, Shi J, Luo P, et al. Spatial as deep: Spatial cnn for traffic scene understanding[C]//Thirty-Second AAAI Conference on Artificial Intelligence. 2018.
3. Lee S, Kim J, Shin Yoon J, et al. Vpgnet: Vanishing point guided network for lane and road marking detection and recognition[C]//Proceedings of the IEEE International Conference on Computer Vision. 2017: 1947-1955.
4. De Brabandere B , Van Gansbeke W , Neven D , et al. End-to-end Lane Detection through Differentiable Least-Squares Fitting[J]. 2019.

1. Ren S, He K, Girshick R, et al. Faster r-cnn: Towards real-time object detection with region proposal networks[C]//Advances in neural information processing systems. 2015: 91-99.
2. Li P, Chen X, Shen S. Stereo R-CNN based 3D Object Detection for Autonomous Driving[J]. arXiv preprint arXiv:1902.09738, 2019.
3. Redmon J, Farhadi A. Yolov3: An incremental improvement[J]. arXiv preprint arXiv:1804.02767, 2018.