Training compact deep learning models for video classification using circulant matrices

TL;DR

This paper introduces compact video classification models using circulant matrices to reduce model size and computational demands, demonstrating effective performance on large-scale datasets.

Contribution

It proposes the use of circulant matrices in state-of-the-art architectures for video classification, achieving a balance between model compactness and accuracy.

Findings

Circulant DBoF achieves high accuracy with reduced model size

Models trained on YouTube-8M dataset show effective size-accuracy trade-offs

Compact models outperform unstructured counterparts in deployment scenarios

Abstract

In real world scenarios, model accuracy is hardly the only factor to consider. Large models consume more memory and are computationally more intensive, which makes them difficult to train and to deploy, especially on mobile devices. In this paper, we build on recent results at the crossroads of Linear Algebra and Deep Learning which demonstrate how imposing a structure on large weight matrices can be used to reduce the size of the model. We propose very compact models for video classification based on state-of-the-art network architectures such as Deep Bag-of-Frames, NetVLAD and NetFisherVectors. We then conduct thorough experiments using the large YouTube-8M video classification dataset. As we will show, the circulant DBoF embedding achieves an excellent trade-off between size and accuracy.