A simple synthetic dataset and baseline model for visual counting. The task is to count the number of even digits given a 100x100 image, each with up to 5 randomly chosen MNIST digits.
We use rejection sampling to ensure digits are separated by at least 28 pixels. Reproduced with details from Learning to count with deep object features.
git clone [email protected]:fomorians/counting-mnist.git
python -m counting_mnist.create_dataset
python -m counting_mnist.main
NOTE: This is not a competitive dataset. It’s intended to be simple place to start for validating ideas in counting models.
We created a lightweight reverse image search and wrote up the details in a three part series for Codementor.
Tool to compute the effective receptive field of a convolutional layer.
A simple synthetic dataset and baseline model for visual counting.
What could you do if your machine learning was actually state-of-the-art?