The DOI matches its Crossref record; 27 of the 32 references it cites were found in a registry or an index.
Generative OpenMax for Multi-Class Open Set Classification
Open accessAbstract
We present a conceptually new and flexible method for multi-class open set classification. Unlike previous methods where unknown classes are inferred with respect to the feature or decision distance to the known classes, our approach is able to provide explicit modelling and decision score for unknown classes. The proposed method, called Gener- ative OpenMax (G-OpenMax), extends OpenMax by employing generative adversarial networks (GANs) for novel category image synthesis. We validate the proposed method on two datasets of handwritten digits and characters, resulting in superior results over previous deep learning based method OpenMax Moreover, G-OpenMax provides a way to visualize samples representing the unknown classes from open space. Our simple and effective approach could serve as a new direction to tackle the challenging multi-class open set classification problem.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“A particularly fruitful approach seems to be the use of Generative Adversarial Networks (GANs) for detecting out-of-sample cases.[24][25][26]”
From The need for uncertainty quantification in machine-assisted medical decision making (2019) · cites this paper as [26]
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 1×
“Driven by recent research in machine learning, the best performing system can achieve impressive results on both small and large scale multi-class classification tasks [8, 9, 13, 21, 28].”
BackgroundImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155 · cited 2×
“Driven by recent research in machine learning, the best performing system can achieve impressive results on both small and large scale multi-class classification tasks [8, 9, 13, 21, 28].”
BackgroundMethodImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333 · cited 4×
“We trained the generator G on ImageNet12 and made several observations: 1) Several classes blend into unrealistic type of object in terms of colour and global structure.”
BackgroundInfluentialImageNet Large Scale Visual Recognition Challenge
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., et al. (2015) · International Journal of Computer Vision · doi:10.1007/s11263-015-0816-y · Cited by 39,380 · cited 1×
“Driven by recent research in machine learning, the best performing system can achieve impressive results on both small and large scale multi-class classification tasks [8, 9, 13, 21, 28].”
BackgroundImage-to-Image Translation with Conditional Adversarial Networks
Isola, P., Zhu, J.-Y., Zhou, T., & Efros, A. A. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.632 · Cited by 22,930 · cited 1×
“GAN has a wide range of applications including, sketch/text to image synthesis [4], image to image translation [10], image editing [19], super resolution [15] and video prediction [6], etc.”
BackgroundAuto-Encoding Variational Bayes
Kingma, D. P., & Welling, M. (2013) · arXiv · doi:10.48550/arxiv.1312.6114 · Cited by 20,067 · cited 2×
“The most prominent are variational autoencoder (VAE) [12] and generative adversarial networks (GAN) [7], among which, GAN is a powerful model of learning arbitrarily complex data distributions.”
BackgroundPhoto-Realistic Single Image Super-Resolution Using a Generative Adversarial Network
Ledig, C., Theis, L., Huszar, F., Caballero, J., Cunningham, A., Acosta, A., Aitken, A., Tejani, A., Totz, J., Wang, Z., et al. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.19 · Cited by 12,696 · cited 1×
“GAN has a wide range of applications including, sketch/text to image synthesis [4], image to image translation [10], image editing [19], super resolution [15] and video prediction [6], etc.”
BackgroundUnsupervised representation learning with deep convolutional generative adversarial networks
Radford, A., Metz, L., & Chintala, S. (2015) · arXiv · doi:10.48550/arxiv.1511.06434 · Cited by 12,235
Improved Techniques for Training GANs
Salimans, T., Goodfellow, I., Zaremba, W., Cheung, V., Radford, A., & Chen, X. (2016) · arXiv · doi:10.48550/arxiv.1606.03498 · Cited by 6,965 · cited 2×
“This is done by using generative adversarial networks (GANs) [7] to generate synthetic samples from unknown classes.”
BackgroundInception-v4, Inception-ResNet and the Impact of Residual Connections on Learning
Szegedy, C., Ioffe, S., Vanhoucke, V., & Alemi, A. (2017) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v31i1.11231 · Cited by 4,636 · cited 1×
“Driven by recent research in machine learning, the best performing system can achieve impressive results on both small and large scale multi-class classification tasks [8, 9, 13, 21, 28].”
Background
Cited by
Sorted by citations
Recent Advances in Open Set Recognition: A Survey
Geng, C., Huang, S.-J., & Chen, S. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.2981604 · Cited by 1,106
“Due to the properties of AL, some researchers also attempt to account for open space with the unknown unknown classes generated by the AL technique [78]–[81].”
BackgroundMethodInfluentialThe need for uncertainty quantification in machine-assisted medical decision making
Begoli, E., Bhattacharya, T., & Kusnezov, D. (2019) · Nature Machine Intelligence · doi:10.1038/s42256-018-0004-1 · Cited by 597
Open Set Domain Adaptation by Backpropagation
Saito, K., Yamamoto, S., Ushiku, Y., & Harada, T. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01228-1_10 · Cited by 603
“Moreover, to give supervision of the unknown samples, a method to generate these samples was proposed [26].”
MethodTowards Open World Object Detection
Joseph, K. J., Khan, S. H., Khan, F. S., & Balasubramanian, V. N. (2021) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr46437.2021.00577 · Cited by 606
“A generative version of OpenMax was proposed in [13] by synthesizing novel class images.”
BackgroundGeneralized Out-of-Distribution Detection: A Survey
Yang, J., Zhou, K., Li, Y., & Liu, Z. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02117-4 · Cited by 884
“[160] [180] [181] [153] [154] [155] [173] [183] [184]”
Open Set Learning with Counterfactual Images
Neal, L., Olson, M., Fern, X. Z., Wong, W., & Li, F. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01231-1_38 · Cited by 510
“We interpolate between the two examples in the latent space as in [7] and apply the generator to the resulting latent point to generate the image:”
BackgroundMethodInfluentialClassification-Reconstruction Learning for Open-Set Recognition
Yoshihashi, R., Shao, W. Q., Kawakami, R., You, S., Iida, M., & Naemura, T. (2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2019.00414 · Cited by 509
“G-Openmax [10], a direct extension of Openmax, trains networks with synthesized unknown data by using generative models.”
BackgroundMethodResultInfluentialC2AE: Class Conditioned Auto-Encoder for Open-Set Recognition
Oza, P., & Patel, V. M. (2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2019.00241 · Cited by 414
“However, these questions are relatively unexplored in the context of deep neural networks [38], [4], [24], [8], [7].”
BackgroundMethodAdversarial Reciprocal Points Learning for Open Set Recognition
Chen, G., Peng, P., Wang, X., & Tian, Y. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3106743 · Cited by 388
“Compared with [22], [28], [37], [38], the main differences of the proposed instantiated adversarial enhancement are as follows: First, the proposed method uses the adversarial mechanism between the close space Sk and the global open space OG formed by the reciprocal points.”
MethodConditional Gaussian Distribution Learning for Open Set Recognition
Sun, X., Yang, Z., Zhang, C., Ling, K. V., & Peng, G. (2020) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr42600.2020.01349 · Cited by 335
“[8] proposed the Generative Openmax method, using generative models to synthesize unknown samples to train the network.”
MethodLearning Placeholders for Open-Set Recognition
Zhou, D.-W., Ye, H.-J., & Zhan, D. (2021) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr46437.2021.00438 · Cited by 268
“We compare to the SOTA methods, i.e ., Softmax, OpenMax [3], G-OpenMax [6], OSRCI [22], C2AE [24], CROSR [36], GFROSR [26]: Softmax : utilizes the highest softmax probability as the confidence score for detection; OpenMax : replaces softmax layer with OpenMax and cal-ibrates the confidence to predict…”
MethodUniversal Source-Free Domain Adaptation
Kundu, J. N., Venkat, N., Rahul, M., & Babu, R. V. (2020) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr42600.2020.00460 · Cited by 286
“In the second kind, Open-set DA [39, 1, 10], the target label space is considered as a superset of the source label space (i.”
BackgroundLearning Open Set Network with Discriminative Reciprocal Points
Chen, G., Qiao, L., Shi, Y., Peng, P., Li, J., Huang, T., Pu, S., & Tian, Y. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58580-8_30 · Cited by 268
“With less information and simple training method, RPL achieves better performance than OSRCI [19] and G-OpenMax [8] using generation-based model on all datasets.”
BackgroundMethodGeneralized Category Discovery
Vaze, S., Hant, K., Vedaldi, A., & Zisserman, A. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.00734 · Cited by 270
“GANs are often employed to generate adversarial samples to train an open-set classifier, e.g, [14, 25, 32].”
BackgroundDeep-Learning-Based Open Set Fault Diagnosis by Extreme Value Theory
Yu, X., Zhao, Z., Zhang, X., Zhang, Q., Liu, Y., Sun, C., & Chen, X. (2021) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2021.3070324 · Cited by 232
OpenGAN: Open-Set Recognition via Open Data Generation
Kong, S., & Ramanan, D. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00085 · Cited by 225
“As a result, prior work in open-set recognition has focused on using GANs to generate realistic open-training images [21, 34, 42].”
InfluentialGenerative-Discriminative Feature Representations for Open-Set Recognition
Perera, P., Morariu, V. I., Jain, R., Manjunatha, V., Wigington, C., Ordóñez, V., & Patel, V. M. (2020) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr42600.2020.01183 · Cited by 209
“In [7], a Generative Adversarial Network (GAN) based framework was used to estimate open-set class activations.”
MethodCounterfactual Zero-Shot and Open-Set Visual Recognition
Yue, Z., Wang, T., Sun, Q., Hua, X., & Zhang, H. (2021) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr46437.2021.01515 · Cited by 221
“The inference methods in OSR calibrate a discriminative model by adjusting the classification logits [5, 75], and the generation methods estimate the seen-class density and reject test samples in the low-density area as outliers [43, 54].”
MethodVisual identification of individual Holstein-Friesian cattle via deep metric learning
Andrew, W., Gao, J., Mullan, S. M., Campbell, N., Dowsey, A. W., & Burghardt, T. (2021) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2021.106133 · Cited by 187
“Alternatives include open-set loss function formulations instead of softmax [69], the generation of counterfactual images close to the training set to strengthen object discrimination [70], and approaches that combine these two techniques [71, 72].”
BackgroundConvolutional Prototype Network for Open Set Recognition
Yang, H.-M., Zhang, X.-Y., Yin, F., Yang, Q., & Liu, C.-L. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.3045079 · Cited by 255
Towards Open Vocabulary Learning: A Survey
Wu, J., Li, X., Xu, S., Yuan, H., Ding, H., Yang, Y., Li, X., Zhang, J., Tong, Y., Jiang, X., et al. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3361862 · Cited by 218
“The main methods used for OSR can be broadly categorized into two types: discriminative models [33], [34], [35], [36], [59] and generative models [60], [61], [62], [63], [64], [65].”
BackgroundMethodTowards Robust Pattern Recognition: A Review
Zhang, X.-Y., Liu, C., & Suen, C. Y. (2020) · Proceedings of the IEEE · doi:10.1109/jproc.2020.2989782 · Cited by 172
Class Anchor Clustering: A Loss for Distance-based Open Set Recognition
Miller, D., Sünderhauf, N., Milford, M. J., & Dayoub, F. (2021) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv48630.2021.00361 · Cited by 174
“This behaviour raises serious concerns about the safety of using CNNs in open set environments [1] – particularly on autonomous systems where per- The authors acknowledge continued support from the Queensland University of Technology (QUT) through the Centre for Robotics.”
BackgroundMethodResultInfluentialOpen-world Machine Learning: Applications, Challenges, and Opportunities
Parmar, J., Chouhan, S. S., Raychoudhury, V., & Rathore, S. S. (2022) · ACM Computing Surveys · doi:10.1145/3561381 · Cited by 163
“[33] Generative OpenMax (G-OpenMax) MNIST and HASYv2 F-measure 80 to 99% for 0 to 13% openness Accuracy 58% (Maximum with MNIST) V.”
BackgroundOpitz, J. (2024) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00675 · Cited by 244
“…when class labels reside on a nominal ‘Likert’ scale (O’Neill, 2017; Amigo et al., 2020), or in a hierarchy (Kosmopoulos et al., 2015), or they are ambiguous and need be matched (e.g., ‘none’/‘null’/‘other’) across annotation styles (Fu et al., 2020), or their number is unknown (Ge et al., 2017).”
BackgroundHybrid Models for Open Set Recognition
Zhang, H., Li, A., Guo, J. J., & Guo, Y. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58580-8_7 · Cited by 151
“Since then, the OpenMax is further developed in [25,6].”
BackgroundMethodInfluentialUnsupervised Out-of-Distribution Detection by Maximum Classifier Discrepancy
Yu, Q., & Aizawa, K. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00961 · Cited by 146
“There are some other studies on open-set classification [1, 2, 6, 21, 23, 24, 30], which involve tasks very similar to OOD detection.”
ResultLearning a Neural-network-based Representation for Open Set Recognition
Hassen, M., & Chan, P. K. (2020) · Society for Industrial and Applied Mathematics eBooks · doi:10.1137/1.9781611976236.18 · Cited by 133
“[12] combine Openmax and GANs[13] for open set recognition.”
BackgroundMethodFew-Shot Open-Set Recognition Using Meta-Learning
Liu, B., Kang, H., Li, H., Hua, G., & Vasconcelos, N. M. (2020) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr42600.2020.00882 · Cited by 128
“We compare the proposed method to several openset recognition SOTA methods, including OpenMax [1], G-OpenMax [5], and Counterfactual [21], and an out-ofdistribution SOTA method, Confidence [14].”
BackgroundMethodInfluentialSpectral-Spatial Latent Reconstruction for Open-Set Hyperspectral Image Classification
Yue, J., Fang, L., & Min, H. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2022.3193747 · Cited by 120
“For the feature extractor corresponding to the open classifier, almost all existing deep open classifiers rely on fully supervised learning [33], [57], [58], [59] on the training set, as shown in Fig.”
MethodZero-Shot Out-of-Distribution Detection Based on the Pre-trained Model CLIP
Esmaeilpour, S., Liu, B., Robertson, E., & Shu, L. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i6.20610 · Cited by 134
Peng, P., Lu, J., Xie, T., Tao, S., Wang, H., & Zhang, H. (2022) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2022.3149935 · Cited by 102
“ForFaults8,12,and14,whichareeasy todetect,ourdetectionratesareallabove95%.”
BackgroundExpanding Low-Density Latent Regions for Open-Set Object Detection
Han, J., Ren, Y., Ding, J., Pan, X., Yan, K., & Xia, G.-S. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.00937 · Cited by 95
“Other approaches include generative adversarial network-based methods [16, 37] which generate potential open-set images to train an open-set classifier, reconstruction-based methods [38, 47, 55] which adopt auto-encoder to recover latent features and identify unknown by reconstruction errors, and…”
BackgroundMethodZhuang, P., Wang, Y., & Qiao, Y. (2018) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3240508.3240616 · Cited by 85
“…Non-Outlier (NNO) [3], etc. Recently, the remarkable successes of convolutional neural networks (CNNs) [9, 18, 20, 24, 37, 38, 45] have promoted open-set deep learning [4, 6, 16, 36] and its relevant applications such as person re-identification [51], face recognition [17], scene parsing [49].”
BackgroundPrototypical Matching and Open Set Rejection for Zero-Shot Semantic Segmentation
Zhang, H., & Ding, H. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00689 · Cited by 81
“Correspond-ingly, some works [40, 52] propose to synthesize images of unknown classes for the training of the network.”
BackgroundMethodInfluentialAdversarial Network with Multiple Classifiers for Open Set Domain Adaptation
(13908942), T. S., (7647290), G. L., (13908870), S. T., (13917924), M. M., & (13908945), F. S. (2021) · Murdoch University Research Portal · Cited by 74
Orientational Distribution Learning with Hierarchical Spatial Attention for Open Set Recognition
Liu, Z., Fu, Y., Pan, Q., & Zhang, Z. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3227913 · Cited by 73
“[22] extended OpenMax by employing generative adversarial networks (GANs).”
BackgroundMethodP-ODN: Prototype-based Open Deep Network for Open Set Recognition
Shu, Y., Shi, Y., Wang, Y., Huang, T., & Tian, Y. (2020) · Scientific Reports · doi:10.1038/s41598-020-63649-6 · Cited by 73
Exploring Rich Semantics for Open-Set Action Recognition
Hu, Y., Gao, J., Dong, J., Fan, B., & Liu, H. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3333206 · Cited by 71
“Recent efforts on OSR have been made towards the following aspects: Generative adversarial network-based methods are utilized to synthesize the samples of the unknown classes to train an open-set classifier [11], [12], [13], [14], [15], [16], [17].”
MethodDeep Learning and Zero-Day Traffic Classification: Lessons Learned From a Commercial-Grade Dataset
Yang, L., Finamore, A., Jun, F., & Rossi, D. (2021) · IEEE Transactions on Network and Service Management · doi:10.1109/tnsm.2021.3122940 · Cited by 69
“In ne r 2017 [35] GAN-OpenMax ML/DL Yes Medium(Weibull inference) 2018 [36] Classifier K+1; ML/DL Yes Low(threshold) 2018 [37]–[40] Clustering loss Functions ML/DL Yes Low(threshold) 2018 [41] Confidence learning ML/DL Yes Low(threshold) 2019 [42] CNN + AE ML/DL None High(AE inference) 2019 [43] AE ML/DL None High(AE…”
BackgroundMethodZheng, S., Chen, S., Yang, L., Zhu, J., Luo, Z., Hu, J., & Yang, X. (2018) · IEEE Access · doi:10.1109/access.2018.2872769 · Cited by 66
“[84] introduced OpenMax layer to estimate the probability of an input being from an unknown class and obtained better performance than”
Few-shot Open-set Recognition by Transformation Consistency
Jeong, M., Choi, S., & Kim, C. (2021) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr46437.2021.01238 · Cited by 64
“At first, we briefly explain the outline of metric-based FSL methods [28, 35] and pseudo-unseen sample-based OSR methods [20, 7, 16].”
BackgroundMethodInfluentialConditional Variational Capsule Network for Open Set Recognition
Guo, Y., Camporese, G., Yang, W., Sperduti, A., & Ballan, L. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00017 · Cited by 64
“Subsequently, [5] extended Openmax to G-Openmax by introducing a generative adversarial network in which the generator produces synthetic samples of novel categories and the discriminator learns the explicit representation for unknown classes.”
BackgroundAdversarial Kinetic Prototype Framework for Open Set Recognition
Xia, Z., Wang, P., Dong, G., & Liu, H. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3231924 · Cited by 59
“Therefore, many studies generated some unknown data and added them into the training phase to further reduce the open space risk, such as [7], [9], [10], [11], and [12].”
BackgroundMethodDifficulty-Aware Simulator for Open Set Recognition
Moon, W., Park, J., Seong, H. S., Cho, C.-H., & Heo, J. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19806-9_21 · Cited by 65
PMAL: Open Set Recognition via Robust Prototype Mining
Lu, J., Xu, Y., Li, H., Cheng, Z., & Niu, Y. F. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i2.20081 · Cited by 5
A Survey on Learning to Reject
Zhang, X.-Y., Xie, G.-S., Li, X., Mei, T., & Liu, C. (2023) · Proceedings of the IEEE · doi:10.1109/jproc.2023.3238024 · Cited by 64
The Importance of Expert Knowledge for Automatic Modulation Open Set Recognition
Li, T., Wen, Z., Long, Y., et al. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3294505 · Cited by 51
Uncertainty for Identifying Open-Set Errors in Visual Object Detection
Miller, D., Sünderhauf, N., Milford, M. J., et al. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3123374 · Cited by 51
“In the open-set classification literature, epistemic uncertainty is obtained by measuring distance in a classifier’s logit space [9], [29]–[31].”
BackgroundCollective Decision of One-vs-Rest Networks for Open-Set Recognition
Jang, J., & Kim, C. O. (2022) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3189996 · Cited by 49
“Furthermore, recently, some researchers have applied autoencoders [12], [15]–[18] and generative adversarial networks (GANs) [19], [20] to provide more discriminative information for unknown identification, in addition to deep neural networks (DNNs) using softmax layers.”
BackgroundInfluentialOpen-Set Classification for Signal Diagnosis of Machinery Sensor in Industrial Environment
Jian-ming, C., Wang, G., Lv, J., He, Z., Yang, T., & Tang, C. (2022) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2022.3169459 · Cited by 49
“In generative models, unknown data are generated from models trained on data of known classes and are used to train the network [15], [16].”
MethodCollective Decision for Open Set Recognition
Geng, C., & Chen, S. (2020) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2020.2978199 · Cited by 50
Task-Adaptive Negative Envision for Few-Shot Open-Set Recognition
Huang, S., Ma, J., Han, G., & Chang, S. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.00703 · Cited by 56
“OR methods typically include class probability re-calibration [1, 21, 36] and negative sample synthesis with generative methods [8, 27].”
MethodA Survey on Open Set Recognition
Mahdavi, A., & Carvalho, M. (2021) · IEEE International Conference on Artificial Intelligence and Knowledge Engineering (AIKE) · doi:10.1109/aike52691.2021.00013 · Cited by 58
“G-OpenMax [15] extends OpenMax in adversarial settings and applies GANs to generate unknown instances.”
Glocal Energy-based Learning for Few-Shot Open-Set Recognition
Wang, H., Pang, G., Wang, P., et al. (2023) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52729.2023.00725 · Cited by 44
Reason and Discovery: A New Paradigm for Open Set Recognition
Fu, Y., Liu, Z., & Lyu, J. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3552760 · Cited by 51
“Ge et al. [8] proposed generative OpenMax (G-Openmax), an extension of Openmax, which arguments the training data with unknown samples generated by a generative adversarial network (GAN) [33].”
InfluentialSpatial location constraint prototype loss for open set recognition
Xia, Z., Wang, P., Dong, G., & Liu, H. (2023) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2023.103651 · Cited by 47
Radar HRRP Open Set Recognition Based on Extreme Value Distribution
Xia, Z., Wang, P., Dong, G., et al. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3257879 · Cited by 41
“[34] proposed the G-OpenMax model by combining the characteristics of generative adversar-”
MethodOpenMix+: Revisiting Data Augmentation for Open Set Recognition
Jiang, G., Zhu, P., Wang, Y., & Hu, Q. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3268680 · Cited by 46
“Similarly, according to their different effects on structural risk (see We focus on negative data augmentation methods which generate negative samples to represent infinite negative space [30], [43], [44], [45], [46].”
BackgroundMethodInfluentialWildfish++: A Comprehensive Fish Benchmark for Multimedia Research
Zhuang, P., Wang, Y., & Qiao, Y. (2020) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2020.3028482 · Cited by 41
Few-Shot Open-Set Recognition of Hyperspectral Images with Outlier Calibration Network
Pal, D., Bundele, V., Sharma, R., Banerjee, B., & Jeppu, Y. (2022) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv51458.2022.00215 · Cited by 43
“Contemporary Open-Set Recognition (OSR) methods apply a threshold on the distance of test samples from the train-3801 ing class prototypes or on the prediction probability post calibrating with a distribution [3, 7, 18].”
MethodOpen-Set Recognition: a Good Closed-Set Classifier is All You Need?Preprint
Vaze, S., Han, K., Vedaldi, A., & Zisserman, A. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.06207 · Cited by 257
Villon, S., Iovan, C., Mangeas, M., & Vigliola, L. (2022) · Sensors · doi:10.3390/s22020497 · Cited by 43
“Applied to DL, most efficient methods rely on “OpenMax”, proposed by [74] and extended by [75,76], which replaces the usual classification layer of deep architecture known as “SoftMax”.”
Active Learning for Open-set Annotation
Ning, K.-P., Zhao, X., Li, Y., & Huang, S.-J. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.00014 · Cited by 44
OpenTAL: Towards Open Set Temporal Action Localization
Bao, W., Yu, Q., & Kong, Y. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.00299 · Cited by 41
“Existing generative OSR methods [13, 18, 21, 33, 47, 68, 74] utilize GAN [23], generative causal model, or mixup augmentation to generate the samples of the unknown.”
MethodGuo, L., Liu, C., Liu, Y., Lin, Y., & Gui, G. (2024) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2024.3408417 · Cited by 38
Open-World Recognition in Remote Sensing: Concepts, challenges, and opportunities
Fang, L., Yang, Z., Ma, T., Yue, J., Xie, W., Ghamisi, P., & Li, J. (2024) · IEEE Geoscience and Remote Sensing Magazine · doi:10.1109/mgrs.2024.3382510 · Cited by 39
“Distinct from generative methods [31], [32], [33], [34], discriminative-based OSR aims to accurately differentiate known classes from unknown ones by identifying appropriate discriminative boundaries [14], [17], [19], [37], [38], [39], [40], [41], [43], [44], [45], [46].”
BackgroundConditional Gaussian Distribution Learning for Open Set RecognitionPreprint
Sun, X., Yang, Z., Zhang, C., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2003.08823 · Cited by 37
“[8] proposed the Generative Openmax method, using generative models to synthesize unknown samples to train the network.”
MethodOpenGAN: Open-Set Recognition via Open Data Generation
Kong, S., & Ramanan, D. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3184052 · Cited by 42
“These additional images are used to augment the training set for learning an open-set model, which oftentimes is designed for both the closed-world K-way classification and open-set discrimination [18, 27, 33].”
InfluentialAl-Dayil, R., Bazi, Y., Alhichri, H. S., et al. (2020) · Remote Sensing · doi:10.3390/rs12111716 · Cited by 36
“[36] presented the Generative OpenMax (G-OpenMax) method.”
Learning Adaptive Embedding Considering Incremental Class
Yang, Y., Sun, Z.-Q., Zhu, H., Fu, Y., Zhou, Y., Xiong, H., & Yang, J. J. (2021) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3109131 · Cited by 39
“utilized the generative model to generate unknown class instances near the decision margin, which can provide explicit probability estimation over generated unknown class [11].”
MethodFu, Y., Liu, Z., & Zhang, Z. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3309091 · Cited by 38
Multi-Task Based Deep Learning Approach for Open-Set Wireless Signal Identification in ISM Band
Gong, J., Qin, X., & Xu, X. (2021) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2021.3118456 · Cited by 35
PSDC: A Prototype-Based Shared-Dummy Classifier Model for Open-Set Domain Adaptation
Liu, Z., Chen, G., Li, Z., Kang, Y., Qu, S., & Jiang, C. (2022) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2022.3228301 · Cited by 34
“Out-of-Distribution Training: For OSR, recent works [40], [51] have shown that outlier exposure, or the ability to train on some outlier instances, can greatly improve open-set discrimination.”
BackgroundFully convolutional open set segmentation
Oliveira, H., Silva, C., Machado, G. L. S., Nogueira, K., & dos Santos, J. A. (2021) · Machine Learning · doi:10.1007/s10994-021-06027-1 · Cited by 37
“While on Vaihingen the did not increase at all in comparison to the Closet Set DNNs with the introduction of OpenPCS, the value for Car had a slight, but noticeable, increase from 0.56 to 0.57.”
BackgroundMethodInfluentialOut-of-Distribution Detection of Human Activity Recognition with Smartwatch Inertial Sensors
Boyer, P., Burns, D., & Whyne, C. M. (2021) · Sensors · doi:10.3390/s21051669 · Cited by 33
“, GAN augmentation may still improve results [47].”
BackgroundOpen Set Recognition in Real World
Yang, Z., Yue, J., Ghamisi, P., Zhang, S., Ma, J., & Fang, L. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02015-9 · Cited by 40
“For instance, Ge Discriminative-based methods generally reject unknown classes by utilizing the latent space features of known classes, employingvariousdiscriminationstrategies(Fig.5illustrates the framework diagram of the discriminative-based OSR).”
Open-Set Human Identification Based on Gait Radar Micro-Doppler Signatures
Ni, Z., & Huang, B. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2021.3052613 · Cited by 32
“Following this direction, [27] proposed the Generative OpenMax (G-OpenMax) algorithm that uses a conditional GAN to generate synthetic samples so as to acquire information about unknown classes.”
MethodGeneralizable Decision Boundaries: Dualistic Meta-Learning for Open Set Domain Generalization
Wang, X., Zhang, J., Qi, L., & Shi, Y. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01062 · Cited by 39
“Other methods [13, 36] adopt generative adversarial networks (GAN) to automatically generate samples of virtual unknown classes.”
MethodOpen World Object Detection: A Survey
Li, Y., Wang, Y., Wang, W., Lin, D., Li, B., & Yap, K. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3480691 · Cited by 50
“Ge et al. proposed the Generative OpenMax (G-OpenMax) [70] to synthesize mixtures of UUCs using a conditional generative adversarial network (GAN) and provide explicit probability estimation over the generated UUCs.”
Sun, H., Yang, B., & Lin, S. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3222494 · Cited by 31
Generative adversarial network based novelty detection usingminimized reconstruction error
Wang, H., Li, X., & Zhang, T. (2018) · Frontiers of Information Technology & Electronic Engineering · doi:10.1631/fitee.1700786 · Cited by 31
“…problems such as clustering (Springenberg, 2015), unsupervised feature learning (Donahue et al., 2016; Dumoulin et al., 2016), classification (Ge et al., 2017), transfer learning (Kim et al., 2017; Yi et al., 2017; Zhu et al., 2017), ensemble learning (Grover and Ermon, 2017), and…”
Learning Relative Feature Displacement for Few-Shot Open-Set Recognition
Deng, S., Yu, J.-G., Wu, Z., Gao, H., Li, Y., & Yang, Y. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3198880 · Cited by 28
“Ge et al. [28] proposed the Generative OpenMax that employs generative adversarial networks for novel category image synthesis.”
BackgroundContrastive Open Set Recognition
Xu, B., Shen, F., & Zhao, J. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i9.26253 · Cited by 35
“In these experiments, we compare our method with the baselines including Softmax Thresholding (Hendrycks and Gimpel 2016), OpenMax (Bendale and Boult 2016), G-OpenMax (Ge et al. 2017), OSRCI (Neal et al. 2018), C2AE (Oza and Patel 2019), RPL++ (Chen et al. 2020a), GFROSR (Perera et al. 2020), PROSER (Zhou, Ye, and Zha…”
BackgroundMethodOpen-Set Radar Emitter Recognition via Deep Metric Autoencoder
Yang, C., Liu, H., Yang, S., et al. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3361899 · Cited by 28
“The existing OSC methods can be categorized as discriminative methods [12], [13], [14], [15], [16] and generative methods [17], [18], [19].”
BackgroundMethodOut-of-Distribution Detection with Semantic Mismatch Under Masking
Yang, Y., Gao, R., & Xu, Q. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20053-3_22 · Cited by 30
Spectral-Spatial Evidential Learning Network for Open-Set Hyperspectral Image Classification
Ji, F., Zhao, W., Wang, Q., Emery, W. J., Peng, R., Man, Y., Wang, G., & Jia, K. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3349415 · Cited by 28
“…unknown classes; moreover, researchers also concentrated on the OSR problem of natural images and achieved satisfactory results [46], [47], [48], [49], where generative-based methods were proven to perform better on unknown class recognition due to a decision boundary recognition ability with…”
BackgroundRD-OpenMax: Rethinking OpenMax for Robust Realistic Open-Set Recognition
Yin, X., Cao, B., Hu, Q., & Wang, Q. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2024.3394890 · Cited by 28
Open-Set Adversarial Defense with Clean-Adversarial Mutual Learning
Shao, R., Perera, P., Yuen, P. C., & Patel, V. M. (2022) · International Journal of Computer Vision · doi:10.1007/s11263-022-01581-0 · Cited by 28
“This idea is further extended by [11] and [27] to exploit a generative model to produce logits of the open-set class.”
BackgroundMoEP-AE: Autoencoding Mixtures of Exponential Power Distributions for Open-Set Recognition
Sun, J., Wang, H., & Dong, Q. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3200112 · Cited by 27
“Existing DNN-based methods can be divided into two main categories: discriminative methods [11], [15], [16], [17], [18], [19], [20], [21], [22], [26], in which representations for open-set recognition are directly learned, and generative methods [3], [8], [23], [24], [25], in which the distributions of features extrac…”
BackgroundMethodMORGAN: Meta-Learning-based Few-Shot Open-Set Recognition via Generative Adversarial Network
Pal, D., Bose, S., Banerjee, B., & Jeppu, Y. (2023) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv56688.2023.00623 · Cited by 27
Multimodal Dual-Embedding Networks for Malware Open-Set Recognition
Guo, J., Wang, H., Xu, Y., Xu, W., Zhan, Y., Sun, Y., & Guo, S. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2024.3373809 · Cited by 29
“For example, the G-OpenMax [40] extends the OpenMax by synthesizing novel unknown samples from the known class data with GANs, such that the classifier can be endowed with some prior knowledge of the distribution or representation of unknown classes.”
OpenGAN: Open Set Generative Adversarial Networks
Ditria, L., Meyer, B. J., & Drummond, T. W. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-69538-5_29 · Cited by 32
Incremental Learning from Low-labelled Stream Data in Open-Set Video Face Recognition
López‐López, E., Pardo, X. M., & Regueiro, C. V. (2022) · Pattern Recognition · doi:10.1016/j.patcog.2022.108885 · Cited by 27
“Meanwhile, generative methods try to provide explicit probability estimation over unknown categories, most of them based on deep networks [11, 37].”
MethodOpen-Set Likelihood Maximization for Few-Shot Learning
Boudiaf, M., Bennequin, E., Tami, M., Toubhans, A., Piantanida, P., Hudelot, C., & Ayed, I. B. (2023) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52729.2023.02299 · Cited by 26
Entropic Open-Set Active Learning
Safaei, B., Vibashan, V. S., de Melo, C. M., & Patel, V. M. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i5.28269 · Cited by 27
Survey on Facial Expression Recognition: History, Applications, and Challenges
Zhao, X., Zhu, J., Luo, B., & Gao, Y. (2021) · IEEE Multimedia · doi:10.1109/mmul.2021.3107862 · Cited by 28
Proportional Similarity-Based Openmax Classifier for Open Set Recognition in SAR Images
Giusti, E., Ghio, S., Oveis, A. H., & Martorella, M. (2022) · Remote Sensing · doi:10.3390/rs14184665 · Cited by 25
Shao, R., Perera, P., Yuen, P. C., & Patel, V. M. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58520-4_40 · Cited by 29
“This formulation was later adopted by [10] and [25] by using a generative model to produce logits of the open-set class.”
BackgroundMethodImage blurring and sharpening inspired three-way clustering approach
Shah, A. A., Azam, N., Alanazi, E., et al. (2022) · Applied Intelligence · doi:10.1007/s10489-021-03072-0 · Cited by 24
“The performance is compared to some notable approaches including DOC, cbsSVM, openMax, 1vs-rest-SVM and others [10, 12, 25, 26].”
Method
Metadata
- Authors
- Ge, Z., Demyanov, S., & Garnavi, R.
- Venue
- Procedings of the British Machine Vision Conference 2017
- Publisher
- British Machine Vision Association
- Published
- 2017
- OpenAlex
- W2963875483
- Cited by
- 472
- References
- 32
- Citations by year
- 2026: 36 · 2025: 59 · 2024: 98 · 2023: 90 · 2022: 63 · 2021: 74 · 2020: 31 · 2019: 12 · 2018: 8 · 2017: 1
- FWCI
- 4.16
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- British Machine Vision Conference (BMVC) (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
- Keywords
- open-set recognition, OpenMax, generative adversarial networks, image generation, unknown class detection, handwritten digits, handwritten characters
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
