Skip to content

AMI-Class: Towards a Fully Automated Multi-view Image Classifier

In this paper, we propose an automated framework for multi-view image classification tasks. We combined a GAN-based multi-view embedding architecture with a scalable AutoML library, DeepHyper. The proposed framework is able to, all at once, train a model to find a common latent representation and perform data imputation, choose the best classifier and tune all necessary hyper-parameters. Experiments on the MNIST data-set show the effectiveness of our solution to optimize the end-to-end multi-view classification pipeline.

Mahmoud Jarraya, Maher Marwani, Gianmarco Aversano, Ichraf Lahouli and Sabri Skhiri, AMI-Class: Towards a Fully Automated Multi-view Image Classifier, In Proc. of The 19th International Conference on Computer Analysis of Images and Patterns CAIP2021, September 2021.

Click here to access the paper.

Share on linkedin
Share on twitter
Share on email

Releated Posts

INTERNSHIPS 2022

This document presents internships supervised by our software engineering department or by our research & development department. Each project is an opportunity to feel both empowered and responsible for your own professional development and for your contribution to the company.
Read More

Multimodal Classifier For Space Target Recognition

We propose a multi-modal framework to tackle the SPARK Challenge by classifying satellites using RGB and depth images. Our framework is mainly based on Auto-Encoders to embed the two modalities in a common latent space in order to exploit redundant and complementary information between the two types of data.
Read More