Developing a car detection program using DNN


I am new in the field of deep learning. I want to develop a program that can detect the cars from a video. Currently, I am doing it through Haar cascade classifier from OpenCV and want to try the same by training my own neural network to improve the accuracy.

Can someone please provide any useful links/ reference to get started with CUDA and DNN on TX1?


Hi, if you would like to train your own networks, you can see our end-to-end tutorial here:

IIRC there is also a vehicle detection demo model included in the Multimedia API samples that JetPack should have installed to your Jetson.


Thanks for the information.

It would be great if the vehicle detection demo model installed by JetPack can do what I want. Could you please tell me where exactly can I find these Multimedia APIs and if this model is accurate enough to detect the cars from a video?

You can find the Multimedia API and samples under your home directory on the Jetson located at ~/tegra_multimedia_api