# DiCENet: Dimension-wise Convolutions for Efficient Networks

## TL;DR

DiCENet introduces a new convolutional unit using dimension-wise convolutions that enhances efficiency and accuracy across multiple vision tasks, outperforming existing models like MobileNetv2 and ShuffleNetv2.

## Contribution

The paper proposes the DiCE unit, a novel convolutional component that improves efficiency and performance when integrated into neural networks, surpassing depth-wise separable convolutions.

## Key findings

- DiCENet achieves 2-4% higher accuracy on ImageNet.
- DiCENet outperforms state-of-the-art models on various vision tasks.
- The DiCE unit is simple, versatile, and improves efficiency across architectures.

## Abstract

We introduce a novel and generic convolutional unit, DiCE unit, that is built using dimension-wise convolutions and dimension-wise fusion. The dimension-wise convolutions apply light-weight convolutional filtering across each dimension of the input tensor while dimension-wise fusion efficiently combines these dimension-wise representations; allowing the DiCE unit to efficiently encode spatial and channel-wise information contained in the input tensor. The DiCE unit is simple and can be seamlessly integrated with any architecture to improve its efficiency and performance. Compared to depth-wise separable convolutions, the DiCE unit shows significant improvements across different architectures. When DiCE units are stacked to build the DiCENet model, we observe significant improvements over state-of-the-art models across various computer vision tasks including image classification, object detection, and semantic segmentation. On the ImageNet dataset, the DiCENet delivers 2-4% higher accuracy than state-of-the-art manually designed models (e.g., MobileNetv2 and ShuffleNetv2). Also, DiCENet generalizes better to tasks (e.g., object detection) that are often used in resource-constrained devices in comparison to state-of-the-art separable convolution-based efficient networks, including neural search-based methods (e.g., MobileNetv3 and MixNet. Our source code in PyTorch is open-source and is available at https://github.com/sacmehta/EdgeNets/