Interpretable CNNs for Object Classification

This paper proposes a generic method to learn interpretable convolutional filters in deep neural network (CNN) for object classification, where each filter encodes features of specific part. Our does not require additional annotations parts or textures supervision. Instead, we use the same training data as traditional CNNs. automatically assigns high conv-layer with an part certain category during learning process. Such explicit knowledge representations conv-layers CNN help people clarify logic encoded CNN, i.e., answering what patterns extracts from input image and uses prediction. We have tested our using different benchmark CNNs various architectures demonstrate broad applicability method. Experiments shown that are much more semantically meaningful than filters.