Improving Shape Awareness and Interpretability in Deep Networks Using Geometric Moments
This addresses the issue of limited shape awareness and interpretability in deep learning for image classification, though it appears incremental as it builds on known mathematical concepts.
The paper tackles the problem of deep networks relying more on texture than shape by introducing a model based on geometric moments to improve shape awareness and interpretability. It achieves higher classification performance compared to baseline and ResNet models on standard datasets.
Deep networks for image classification often rely more on texture information than object shape. While efforts have been made to make deep-models shape-aware, it is often difficult to make such models simple, interpretable, or rooted in known mathematical definitions of shape. This paper presents a deep-learning model inspired by geometric moments, a classically well understood approach to measure shape-related properties. The proposed method consists of a trainable network for generating coordinate bases and affine parameters for making the features geometrically invariant yet in a task-specific manner. The proposed model improves the final feature's interpretation. We demonstrate the effectiveness of our method on standard image classification datasets. The proposed model achieves higher classification performance compared to the baseline and standard ResNet models while substantially improving interpretability.