Here's an animated example of what 2D convolution might look like: http://deeplearning.net/software/theano/tutorial/conv_arithmetic.html
ravenseattuna
Convolution also plays a big part in object detection! By extracting the object you want to detect's key features (for example, the many curves in the number 8), you can create a filter that detects these curves (usually just a smaller nxn matrix that represents some curves) and, by convolving over the image you want to detect objects on, see whether the returned image contains these features, due to larger than usual values existing int he final matrix! A link explaining this is here: https://adeshpande3.github.io/A-Beginner%27s-Guide-To-Understanding-Convolutional-Neural-Networks/
Here's an animated example of what 2D convolution might look like: http://deeplearning.net/software/theano/tutorial/conv_arithmetic.html
Convolution also plays a big part in object detection! By extracting the object you want to detect's key features (for example, the many curves in the number 8), you can create a filter that detects these curves (usually just a smaller nxn matrix that represents some curves) and, by convolving over the image you want to detect objects on, see whether the returned image contains these features, due to larger than usual values existing int he final matrix! A link explaining this is here: https://adeshpande3.github.io/A-Beginner%27s-Guide-To-Understanding-Convolutional-Neural-Networks/