CNN Concepts
Convolution, Padding, Stride and Pooling
How does a Convolution layer work?
• We use multiple convolution filters or kernels that run over the image
• compute a dot product. Each filter extracts different features from the image.
• Lets consider a filter of size 3x3 and an image of size 5x5. This provides us a single
value for the feature cell.
How does a Convolution layer work?
Sliding Patch
Stride
Stride
Padding
Padding
Pooling
Pooling
Max Pooling
Avg Pooling
• Take Average in place of Maximum