Conditional Gaussian Distribution Learning for Open Set Recognition

Abstract

Deep neural networks have achieved state-of-the-art performance in a wide range of recognition/classification tasks. However, when applying deep learning to real-world applications, there are still multiple challenges. A typical challenge is that unknown samples may be fed into the system during the testing phase and traditional deep neural networks will wrongly recognize the unknown sample as one of the known classes. Open set recognition is a potential solution to overcome this problem, where the open set classifier should have the ability to reject unknown samples as well as maintain high classification accuracy on known classes. The variational auto-encoder (VAE) is a popular model to detect unknowns, but it cannot provide discriminative representations for known classification. In this paper, we propose a novel method, Conditional Gaussian Distribution Learning (CGDL), for open set recognition. In addition to detecting unknown samples, this method can also classify known samples by forcing different latent features to approximate different Gaussian models. Meanwhile, to avoid information hidden in the input vanishing in the middle layers, we also adopt the probabilistic ladder architecture to extract high-level abstract features. Experiments on several standard image datasets reveal that the proposed method significantly outperforms the baseline method and achieves new state-of-the-art results.

arXiv commentAccepted to CVPR2020

References in corpus (4)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Auto-Encoding Variational Bayes
• LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop
• Semi-Supervised Learning with Ladder Networks

Cited by in corpus (10)

• Collective Decision of One-vs-Rest Networks for Open Set Recognition
• Inspect, Understand, Overcome: A Survey of Practical Methods for AI Safety
• Open-Set Recognition: a Good Closed-Set Classifier is All You Need?
• DeepEMD: Differentiable Earth Mover's Distance for Few-Shot Learning
• M2IOSR: Maximal Mutual Information Open Set Recognition
• Open Set Recognition with Conditional Probabilistic Generative Models
• Few-shot Open-set Recognition by Transformation Consistency
• Synthetic Unknown Class Learning for Learning Unknowns
• Calibrating Class Activation Maps for Long-Tailed Visual Recognition
• One-Class Meta-Learning: Towards Generalizable Few-Shot Open-Set Classification