Learning $3$D-FilterMap for Deep Convolutional Neural Networks

TL;DR

This paper introduces a new compact CNN architecture called 3D-FilterMap CNNs that uses shared 3D filter representations to reduce parameters while maintaining performance.

Contribution

The paper proposes a novel 3D-FilterMap approach for CNNs that learns compact filter representations with weight sharing, differing from network compression methods.

Findings

Reduces parameter size significantly compared to traditional CNNs.

Maintains comparable performance with baseline CNNs.

Demonstrates effectiveness on standard benchmarks.

Abstract

We present a novel and compact architecture for deep Convolutional Neural Networks (CNNs) in this paper, termed $3$D-FilterMap Convolutional Neural Networks ($3$D-FM-CNNs). The convolution layer of $3$D-FM-CNN learns a compact representation of the filters, named $3$D-FilterMap, instead of a set of independent filters in the conventional convolution layer. The filters are extracted from the $3$D-FilterMap as overlapping $3$D submatrics with weight sharing among nearby filters, and these filters are convolved with the input to generate the output of the convolution layer for $3$D-FM-CNN. Due to the weight sharing scheme, the parameter size of the $3$D-FilterMap is much smaller than that of the filters to be learned in the conventional convolution layer when $3$D-FilterMap generates the same number of filters. Our work is fundamentally different from the network compression literature that reduces the size of a learned large network in the sense that a small network is directly learned from scratch. Experimental results demonstrate that $3$D-FM-CNN enjoys a small parameter space by learning compact $3$D-FilterMaps, while achieving performance compared to that of the baseline CNNs which learn the same number of filters as that generated by the corresponding $3$D-FilterMap.