GTSRB Challenge

a.k.a German Traffic Sign Recognition Benchmark :de: :no_entry: :no_bicycles: :no_entry_sign: ...

Goal

Use Torch to train and evaluate a 2-stage convolutional neural network able to classify German traffic sign images (43 classes):

• fork the repository under your account,
• go to Settings > Features and enable Issues,
• create an issue under your repo describing your approach,
• report your result(s),
• commit your code,
• edit the README with pre-requisites and usage,
• boost accuracy by experimenting the multi-scale architecture,
• compare with the results obtained in matching mode (i.e use the features with a distance-based search).

Repo

Info

This repository has been forked from the Moodstocks/gtsrb repository. It contains the code used to approach the GTSRB challenge as suggest in "Goal" above.

Execution

To execute this code,

• Download the GTSRB dataset using the "download.sh" script:

  > ./download.sh

• Execute one of the "run-*.lua" scripts using torch7. Please check the scripts to see the available options. Example:

  > th run-cnnDropOut2.lua -save results-cnn-do2 -optimization SGD -learningRate 1e-3 -learningRateDecay 1e-7 -momentum 0.9 -plot

To test the results after a given iteration (here iteration 5), use the script "scripts/eval.lua" as follow:

> th scripts/eval.lua results-cnn-do2/hyp_epoch5.csv data/GT-final_test.csv

Paper

Traffic Sign Recognition with Multi-Scale Convolutional Networks, by Yann LeCun et al.

Dataset

Training

http://benchmark.ini.rub.de/Dataset/GTSRB_Final_Training_Images.zip (263 MB)

Testing

http://benchmark.ini.rub.de/Dataset/GTSRB_Final_Test_Images.zip (84 MB) http://benchmark.ini.rub.de/Dataset/GTSRB_Final_Test_GT.zip (98 kB)