I was wondering if it is possible use HoG features to train a Neural Network, I know that in the original paper by Dalal and Triggs they used the data generated to train a SVM.

If not is possible or suitable do you know a descriptor to get features in order to train a neural network.


1 Answer 1


Neural networks work on any data with consistent dimenionality. If your images all have the same size, than the HOG features will also have consistent dimensionality over all images.

Indeed it is quite standard to train NNs on complexer features like HOG or SIFT instead of on the raw pixel data.

As a deeper point, HOG can be computed using NNs. In fact, the HOG computation can be seen as a specific instance of convolutional neural networks, albeit with some obscure connectivity restrictions and obscure kernel functions (like atan, and normalization functions). This is a purely theoretical point, however - I would not recommend implementing HOG using NNs!


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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