For a current project demo, I'm searching for a car detection neural network in Python written in TF/Keras (or any other type, as long as it has no C++ dependencies). Later on, I gonna write my own, but, for now, I need a working substitute. It should be easy to use and implement in my own Python code, even better if it's already trained. I searched the whole GitHub for this, but I can't find anything that would fit my needs.

Does someone know where to find this type of code? And are there ones for pedestrian detection, too?

  • $\begingroup$ just search for object detection networks, there's plenty of them (e.g. YOLOv3, R-CNN, SSD), they will be able to detect cars, pedestrians and many more. You can later train specific architecture just to detect cars and pedestrians if you want. $\endgroup$ – Brale Feb 23 at 22:27
  • $\begingroup$ Thanks for the answer, but this is what I did so far and didn't found any that would be good for me. Can you link something specific, e.g. one of that networks or a place where I can search other than Github? $\endgroup$ – Tamas Nemes Feb 24 at 12:09
  • $\begingroup$ I don't know what does "good for me" mean. Like I said you can find them easily, I found 4 by googling in a minute. You have some of them here, here , here, here $\endgroup$ – Brale Feb 24 at 12:58

I recall seeing several data sets on Kaggle on the topic of car images. Usually these data sets have associated with them user uploaded python kernels that do either classification( what make of car) or detection ( show a bounding box on the car). I would recommend you go to Kaggle then select data sets and do a search for 'cars'. I did a quick look and found a data set at https://www.kaggle.com/jessicali9530/stanford-cars-dataset that has 18 kernels associated with it. There well may be many more data sets available. Good luck in your search!

| improve this answer | |

Not the answer you're looking for? Browse other questions tagged or ask your own question.