NITCAD-dataset

As part of our major project, we have worked on to develop a dataset for autonomous vehicles in India (named it as NITCAD i.e National Insititute of Technology Calicut Autonomous Driving).

Please refer to NITCAD dataset paper to understand more about the dataset. Alternatively you can read this Medium Blog for basic introduction to dataset. Fill this Google form link (also attached in blogpost) to get access to this dataset.

If this work helps in your research, please consider to cite as

@article{srinath2020nitcad,
  title={NITCAD-Developing an object detection, classification and stereo vision dataset for autonomous navigation in Indian roads},
  author={Srinath, Namburi GNVV Satya Sai and Joseph, Athul Zac and Umamaheswaran, S and Priyanka, Ch Lakshmi and Nair, Malavika and Sankaran, Praveen},
  journal={Procedia Computer Science},
  volume={171},
  pages={207--216},
  year={2020},
  publisher={Elsevier}
}

Also you can refer to Speed estimation using Stereo Vision images for the speed estimation by using SIFT (Scale Invariant Feature Transform), YOLO and MC-CNN.

If this work helps in your research, please consider to cite as

@inproceedings{umamaheswaran2019stereo,
  title={Stereo Vision Based Speed Estimation for Autonomous Driving},
  author={Umamaheswaran, S and Nair, Malavika and Joseph, Athul Zac and Srinath, Namburi GNVV Satya Sai and Priyanka, Ch Lakshmi and Sankaran, Praveen},
  booktitle={2019 International Conference on Information Technology (ICIT)},
  pages={201--205},
  year={2019},
  organization={IEEE}
}

In addition to that, we worked on KITTI dataset and here are few algorithms that we worked

KITTI LiDAR dataset

1. Gaussian Mixture Model
2. Fast Segmentation
3. Connected Component Labelling

Object detection

1. Faster R-CNN
2. YOLO

Depth estimation by Stereo Vision based approach

1. MC-CNN (Matching Cost CNN)
2. Inbuild functions from OpenCV

Monocular Visual Odometry (Kanade–Lucas–Tomasi feature tracker)

Object recognition by DenseNet, Inceptionv3, MobileNet, NASNet, VGG16 and Xception (Keras implementation)

License

NITCAD - An object detection, classification and stereo vision dataset for autonomous navigation in Indian roads by Namburi GNVV Satya Sai Srinath, Athul Zac Joseph, S Umamaheswaran, Ch. Lakshmi Priyanka, Malavika Nair M, Praveen Sankaran is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. Based on a work at https://www.sciencedirect.com/science/article/pii/S187705092030987X?via%3Dihub%3C/a%3E

I would like to thank my project teammates Athul Zac Joseph, Ch. Lakshmi Priyanka, Malavika Nair M, S Umamaheswaran and our guide Dr. Praveen Sankaran who helped at various stages during the project.

Note: This repository is in progress and additional links/data will be added.

Pending tasks

• Add odometry videos
• Add codes related to deep learning (object classification)