Deep CNN-based Multi-task Learning for Open-Set Recognition

TL;DR

This paper introduces a deep CNN multi-task learning framework that enhances open-set visual recognition by combining classification and reconstruction, utilizing statistical modeling of errors for improved detection of unknown classes.

Contribution

It presents a novel multi-task CNN architecture that integrates classification and decoding with error distribution modeling for better open-set recognition.

Findings

Significantly outperforms existing open-set recognition methods.

Utilizes reconstruction error and EVT for effective unknown class detection.

Demonstrates robustness across multiple image datasets.

Abstract

We propose a novel deep convolutional neural network (CNN) based multi-task learning approach for open-set visual recognition. We combine a classifier network and a decoder network with a shared feature extractor network within a multi-task learning framework. We show that this approach results in better open-set recognition accuracy. In our approach, reconstruction errors from the decoder network are utilized for open-set rejection. In addition, we model the tail of the reconstruction error distribution from the known classes using the statistical Extreme Value Theory to improve the overall performance. Experiments on multiple image classification datasets are performed and it is shown that this method can perform significantly better than many competitive open set recognition algorithms available in the literature. The code will be made available at: github.com/otkupjnoz/mlosr.