# Slim-CNN: A Light-Weight CNN for Face Attribute Prediction

## TL;DR

This paper presents Slim-CNN, a lightweight and efficient neural network architecture for face attribute prediction that maintains high accuracy with significantly fewer parameters, suitable for mobile devices.

## Contribution

Introduction of Slim Modules combining depthwise and pointwise convolutions to create a compact, high-accuracy CNN for face attribute prediction.

## Key findings

- Achieves 91.24% accuracy on CelebA dataset.
- Uses at least 25 times fewer parameters than comparable methods.
- Reduces memory storage by at least 87%.

## Abstract

We introduce a computationally-efficient CNN micro-architecture Slim Module to design a lightweight deep neural network Slim-Net for face attribute prediction. Slim Modules are constructed by assembling depthwise separable convolutions with pointwise convolution to produce a computationally efficient module. The problem of facial attribute prediction is challenging because of the large variations in pose, background, illumination, and dataset imbalance. We stack these Slim Modules to devise a compact CNN which still maintains very high accuracy. Additionally, the neural network has a very low memory footprint which makes it suitable for mobile and embedded applications. Experiments on the CelebA dataset show that Slim-Net achieves an accuracy of 91.24% with at least 25 times fewer parameters than comparably performing methods, which reduces the memory storage requirement of Slim-net by at least 87%.