Generative-Discriminative Feature Representations for Open-Set Recognition

Perera, Pramuditha; Morariu, Vlad I.; Jain, Rajiv; Manjunatha, Varun; Wigington, Curtis; Ordóñez, Vicente; Patel, Vishal M.

doi:10.1109/cvpr42600.2020.01183

Cited by 154 publications

(128 citation statements)

References 13 publications

Supporting

Mentioning

126

Contrasting

Order By: Relevance

“…This shows that the network structure and the inference strategy introduced in the paper has contributed towards improving open-set recognition performance of the network. At the same time we note that the proposed network has used less than 10% of the parameters that recent open-set networks have used [7], [6], [8], [9], [10]. In Table 2, we tabulate the closed set accuracy for known classes corresponding to the proposed method and the data-aug baseline.…”

Section: Resultsmentioning

confidence: 99%

“…On the other hand, when the detector declares a query to be a known class instance, it is passed through the classifier to determine the class label. Recent works in open-set recognition has focused on modeling known class concepts [8], [9], [10] or simulating open-set samples [6], [7]. Both of these strategies take advantage of external deep networks to carry out training.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Geometric Transformation-Based Network Ensemble for Open-Set Recognition

Perera

Patel

2021

2021 IEEE International Conference on Multimedia and Expo (ICME)

Self Cite

View full text Add to dashboard Cite

Open-set recognition focuses on the problem of determining whether a given query image belongs to one of the classes known to the network. Recent works in open-set recognition have attempted to solve this problem using an external deep network by either modeling known class samples or simulating open-set samples at the expense of more parameters. In this work, we propose a modified network structure and an inference rule that leads to better open-set recognition performance. First, we show that networks learn different representations when they are trained on datasets subjected to extreme geometric transformations. By exploiting this fact, the proposed mechanism learns multiple representations from the same set of image data using a set of parallel networks with identical structures. During inference, decisions of all independent networks are fused using majority voting to arrive at predictions. The proposed method obtains state-of-the-art open-set detection performance on multiple object recognition datasets.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Geometric Transformation-Based Network Ensemble for Open-Set Recognition

Perera

Patel

2021

2021 IEEE International Conference on Multimedia and Expo (ICME)

Self Cite

View full text Add to dashboard Cite

show abstract

“…• In OSR, information can only be extracted from known samples. To realize unknown detection, some OSR methods are based on generative models [9] [10] [4] [11] [12] (such as Generative Adversarial Networks (GAN) [13] or Variational Chapter 1. Introduction 3 Auto-Encoders (VAE) [14]).…”

Section: Challenges and Motivationsmentioning

confidence: 99%

“…• Class-specific features can help open set classifiers categorize known classes as well as identify unknown samples during testing [12]. AE models are well known to extract class-specific features through a pixel-level reconstruction strategy, where each pixel in the reconstructed image is expected to be exactly the same as the input image.…”

Section: Challenges and Motivationsmentioning

confidence: 99%

See 1 more Smart Citation

A study on open set recognition methods

Sun¹

View full text Add to dashboard Cite

Ling Keck-Voon. He constantly encouraged and supported me throughout my Ph.D. career. He has helped me a lot in how to do academic research, in-depth thinking, and manuscript writing. I would like to thank Professor Lin Guoshen for his patient guidance and for leading me into the wonderful world of deep learning. Also, I would like to thank Nanyang Technological University, National Research Foundation Singapore, and SMRT Corporation for funding my education and research.

show abstract

Open-Set Adversarial Defense

Shao

Perera

Yuen

et al. 2020

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Previous studies have shown the vulnerability of vision transformers to adversarial patches, but these studies all rely on a critical assumption: the attack patches must be perfectly aligned with the patches used for linear projection in vision transformers. Due to this stringent requirement, deploying adversarial patches for vision transformers in the physical world becomes impractical, unlike their effectiveness on CNNs. This paper proposes a novel method for generating an adversarial patch (G-Patch) that overcomes the alignment constraint, allowing the patch to launch a targeted attack at any position within the field of view. Specifically, instead of directly optimizing the patch using gradients, we employ a GAN-like structure to generate the adversarial patch. Our experiments show the effectiveness of the adversarial patch in achieving universal attacks on vision transformers, both in digital and physical-world scenarios. Additionally, further analysis reveals that the generated adversarial patch exhibits robustness to brightness restriction, color transfer, and random noise. Real-world attack experiments validate the effectiveness of the G-Patch to launch robust attacks even under some very challenging conditions.

show abstract

Generative-Discriminative Feature Representations for Open-Set Recognition

Cited by 154 publications

References 13 publications

Geometric Transformation-Based Network Ensemble for Open-Set Recognition

Geometric Transformation-Based Network Ensemble for Open-Set Recognition

A study on open set recognition methods

Open-Set Adversarial Defense

Contact Info

Product

Resources

About