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

By Ankit Kumar Sharma, Hassan Foroosh. 2019

52 citations recorded by the source corpus.

We introduce a computationally-efficient CNN micro-architecture Slim Module to design a lightweight deep neural network, Slim-CNN, 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 multiple Slim Modules to devise a compact CNN, which still maintains very high accuracy. Additionally, Slim-CNN has a very low memory footprint, which makes it suitable for mobile and embedded applications. Experiments on the CelebA dataset show that Slim-CNN achieve an accuracy of 91.24% with 25x fewer parameters compared to MCNN-AUX and 100x fewer parameters when compared to DTML. This reduces the memory storage requirement of Slim-CNN by at least 87%. Furthermore, we compare Slim Modules with other well-known micro-architectures, such as Inception modules, residual blocks, Shuffle-Unit, and Inverted Residual units, and show it outperforms them in performance and in memory size, making it suitable for face-related tasks in embedded applications.

Subfields: Computer Vision and Pattern Recognition, Efficiency & Compression

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Efficiency & Compression

Citations in the corpus

Rewriting a Deep Generative Model (2020, methodology): “For the smiling edit, we measure eﬃcacy by counting images classiﬁed as smiling by an attribute classiﬁer [69], and we also quantify changes made in the images outside the mouth region by masking lips using a face segmentation model [87] and using LPIPS [84] to quantify changes.”