You seek to know more about computer vision through CNN applications?
Deep neural networks have tremendous potential to learn complex non-linear functions, patterns, and representations. This includes real-world applications like image categorization and classification and the very popular concept of image artistic style transfer. Computer vision is all about the art and science of making machines understand high-level useful patterns and representations from images and videos so that it would be able to make intelligent decisions similar to what a human would do upon observing its surroundings.
Convolutional neural networks or CNNs are extensively used for automated feature extraction in images. In fact, CNNs are similar to the general deep neural networks, but with explicit assumption of input being a data set where which the location of a feature is relevant can be attempted via CNNs like image, but not limited to then.
I shared a Notebook on github that explore convolutional neural networks through the task of image classification using publicly dataset CIFAR-10, to:
- Image classification use CNNs from scratch
- Transfer learning: image classification using pre-trained models
With the desire to have a reference for similar cases, this notebook has a summary of the concepts and methods applied, as links to some more detailed references. I hope the same one helps somebody else.
If you prefer, you can access the same Notebook on a on Kaggle kernel.
