Gram Regularization for Multi-view 3D Shape Retrieval

TL;DR

This paper introduces Gram regularization, a novel method to improve 3D shape retrieval by enhancing neural network generalization and discriminative feature extraction, outperforming traditional regularization techniques.

Contribution

The paper proposes Gram regularization, a new regularization term that encourages diverse weight kernels, improving 3D shape retrieval performance and can be integrated into existing architectures.

Findings

Gram regularization improves retrieval accuracy on ModelNet.

It converges quickly and is easy to implement.

Outperforms L2 regularization in experiments.

Abstract

How to obtain the desirable representation of a 3D shape is a key challenge in 3D shape retrieval task. Most existing 3D shape retrieval methods focus on capturing shape representation with different neural network architectures, while the learning ability of each layer in the network is neglected. A common and tough issue that limits the capacity of the network is overfitting. To tackle this, L2 regularization is applied widely in existing deep learning frameworks. However,the effect on the generalization ability with L2 regularization is limited as it only controls large value in parameters. To make up the gap, in this paper, we propose a novel regularization term called Gram regularization which reinforces the learning ability of the network by encouraging the weight kernels to extract different information on the corresponding feature map. By forcing the variance between weight kernels to be large, the regularizer can help to extract discriminative features. The proposed Gram regularization is data independent and can converge stably and quickly without bells and whistles. Moreover, it can be easily plugged into existing off-the-shelf architectures. Extensive experimental results on the popular 3D object retrieval benchmark ModelNet demonstrate the effectiveness of our method.