A novel network training approach for open set image recognition

TL;DR

This paper introduces OSRNet, a new deep learning approach for open set image recognition that uses a mined 'Known UnknownTrainer' dataset to improve detection of unknown classes while maintaining accuracy on known classes.

Contribution

The paper proposes a novel method for mining a 'Known UnknownTrainer' set and designing OSRNet, a deep network that effectively detects unknown classes in open set recognition tasks.

Findings

OSRNet outperforms existing methods on six benchmark datasets.

The approach effectively detects unknown classes while preserving accuracy on known classes.

Mining the KUT set improves open set recognition performance.

Abstract

Convolutional Neural Networks (CNNs) are commonly designed for closed set arrangements, where test instances only belong to some "Known Known" (KK) classes used in training. As such, they predict a class label for a test sample based on the distribution of the KK classes. However, when used under the Open Set Recognition (OSR) setup (where an input may belong to an "Unknown Unknown" or UU class), such a network will always classify a test instance as one of the KK classes even if it is from a UU class. As a solution, recently, data augmentation based on Generative Adversarial Networks(GAN) has been used. In this work, we propose a novel approach for mining a "Known UnknownTrainer" or KUT set and design a deep OSR Network (OSRNet) to harness this dataset. The goal isto teach OSRNet the essence of the UUs through KUT set, which is effectively a collection of mined "hard Known Unknown negatives". Once trained, OSRNet can detect the UUs while maintaining high classification accuracy on KKs. We evaluate OSRNet on six benchmark datasets and demonstrate it outperforms contemporary OSR methods.