← All projects

Traffic Sign Recognition: What Each CNN Component Buys You

A controlled ablation on the German Traffic Sign Recognition Benchmark (43 classes, IJCNN 2011). Start from a plain CNN, change one thing at a time, and measure what it does to generalisation.

  • Ablation study
  • Neural Networks & Deep Learning course
  • University of Tehran · 2020
  • Keras · GTSRB
  • with Fatemeh Haghighi
German traffic signs
43traffic-sign classes
26,640training images per epoch
91.8% → 96.8%validation accuracy, baseline to batch-norm model
5single-factor experiments

Explore the ablations

One change at a time

Interactive

Final validation accuracy of each run, read from the training logs in the notebook. Pick an experiment.

Takeaways

  1. The baseline overfitsIt reaches 100% training accuracy within 25 epochs, while validation stalls near 92%: capacity is not the problem, generalisation is.
  2. Regularisation pays mostDropout(0.5) lifts validation accuracy from 92.5% to 96.5% at the same 100 epochs.
  3. Batch norm trains fastestWith batch norm the model hits 96.8% in 30 epochs; the same network without it is at 78.6%.
  4. Augmentation can hurtRandom transforms made validation loss unstable and cost 8 points here. A likely cause: some transforms change a sign's meaning (a mirrored "turn left" reads as "turn right").