Skip to content

edras/tucson-backseat

Repository files navigation

tucson-backseat

Technical Challenge for DL on dataset from https://sviro.kl.dfki.de/data/

you will need to download and extract these datasets:

you also need environment configured with:

  • tensorflow
  • keras
  • opencv
  • scikit-learn
  • numpy
  • imutils

to train the network, execute command:

python train_model.py --dataset tucson_RGB/tucson/train/RGB --output output

to test the trained model, execute command:

python test_model.py --input tucson_RGB_wholeImage/tucson/test/RGB_wholeImage --model output/weights-028-0.0554.hd

To see the results and some description, check 'Challenge.pdf'.

About

Technical Challenge for DL on dataset from https://sviro.kl.dfki.de/data/

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published