A Boundary Based Out-of-Distribution Classifier for Generalized Zero-Shot Learning

Chen, Xingyu; Lan, Xuguang; Sun, Fuchun; Zheng, Nanning

doi:10.1007/978-3-030-58586-0_34

Cited by 49 publications

(31 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For specific OoC per class in SIXray and CIFAR-10, we observe that when using REFGAN and REFGAN BOUNDARY, classes which were misdetected using the PRIOR, now have improved performance. In the future, we will evaluate REFGAN using a large number of classes using (9) to avoid softmax [39] and compare it to [15] and meta-learning methods that address the few-shot learning setting [25], [40], [41]. In addition to fewshot OoD detection, our algorithm will also output confidence and bounding boxes around anomalies/objects within a robust integrated classification-with-OoD-detection framework.…”

Section: Discussionmentioning

confidence: 99%

Few-Shot Adaptive Detection of Objects of Concern Using Generative Models with Negative Retraining

Dionelis

Vaighan

Tsaftaris

2021

2021 IEEE 33rd International Conference on Tools With Artificial Intelligence (ICTAI)

View full text Add to dashboard Cite

Detecting objects which we are interested in, Objects of Concern (OoC), is nowadays attracting attention. In aviation and transport, it is important to robustly detect OoC in security images. OoC are rare, differ from typical samples, and may be unknown during training. Most OoC detection methods need to be trained on large datasets to achieve good performance and have limited real-world generalization ability. A large variety of samples is needed, and it is expensive to collect and label large datasets due to the rarity of OoC. To address such limitations, we propose the negative REtraining with Few-shots Generative Adversarial Network (REFGAN) for detecting OoC. REFGAN aims at automatically identifying OoC by learning from Objects of No Concern (OoNC) and OoC. Our methodology comprises learning a prior using OoNC, and few-shot adaptation using the Few-Shot OoC (FSOoC). We propose a methodology to robustly perform few-shot adaptive detection of OoC using GANs and learned distribution boundaries. The evaluation of REFGAN on the Baggage SIXray dataset shows that when FSOoC are used, our model outperforms the prior improving OoC detection, and outperforms recent benchmarks by approximately 6.3% in mean values. REFGAN using few-shots of 80 samples shows a robust comparable performance compared to REFGAN using all the samples for retraining and model adaptation. REFGAN can detect unknown OoC and its evaluation on SIXray and CIFAR-10 shows robustness against the number of few-shot samples of OoC. REFGAN on CIFAR-10 outperforms benchmarks by approximately 16% using few-shots of 80 and of 10 samples.

show abstract

Section: Discussionmentioning

confidence: 99%

Few-Shot Adaptive Detection of Objects of Concern Using Generative Models with Negative Retraining

Dionelis

Vaighan

Tsaftaris

2021

2021 IEEE 33rd International Conference on Tools With Artificial Intelligence (ICTAI)

View full text Add to dashboard Cite

show abstract

“…In GZSL setting, OoD samples indicate samples from unseen categories, and In-Domain (ID) samples indicate samples from seen categories. Therefore, these OoD based GZSL approaches [8] firstly try to differentiate the test sample whether it belongs to OoD samples, and then use a seen classes classifier usually Softmax for ID samples and a ZSL classifier such as ALE [2] for OoD samples respectively. Besides, Atzmon et al proposed a soft gating model which makes soft decisions if a sample is from a seen class [7].…”

Section: Related Workmentioning

confidence: 99%

“…However, direct search within all classes cannot well utilized the knowledge learned from the seen training samples, thus some Out-of-Domain (OoD) based methods are proposed to first classify the feature as seen or unseen, and then divide the GZSL problem into two sub-tasks: a conventional ZSL task and a fully supervised learning task. For example, some OoD [7,8] based methods define two classifiers to handle the seen and unseen domains separately. However, they all neglected that OoD detection is also a binary zeroshot classification, so it is unsuitable to use two totally different models for OoD detection and zero-shot classification respectively.…”

Section: Introductionmentioning

confidence: 99%

Dual Prototype Relaxation for Generalized Zero Shot Learning

Zhang

Hu³

2021

2021 IEEE International Conference on Multimedia and Expo (ICME)

View full text Add to dashboard Cite

Generalized Zero Shot Learning (GZSL) is proposed to solve the training data missing problem by transferring the knowledge learned in seen classes to unseen classes. Many methods project the visual features into semantic space and find their nearest neighbours among the pre-defined attributes, which has achieved significant success. However, there are two problems involved in this type of methods, one is that the projection is a many-to-one mapping, which cannot maintain the diversity of features in semantic space, the other is that searching within all classes cannot well utilize the knowledge learned in the seen classes. In this paper, we propose a novel method named Dual Prototype Relaxation (DPR) by relaxing the projection from many-to-one to many-to-many. Specifically, we add noise to the semantic prototype in response to the projection of multiple features within a class, and reconstruct the visual features with the same relaxed prototype. Besides, in order to make better use of the knowledge learned in the seen classes, an Out-of-Domain (OoD) based method is employed to first classify the feature to seen or unseen domains, and then the same DPR model is applied to recognize its category within each domain. Extensive experiments on four popular datasets are conducted and the results show that our method can outperform many linear and deep state-ofthe-art methods although our method is a linear one.

show abstract

“…Since only seen domain images are available during training, images from unseen categories tend to be recognized as seen categories. To this end, DVBE [31] and Boundary-based OOD [10] explore out-of-distribution detection to treat seen and unseen domains separately. Some works [28,20] suppress the seen category confidence when recognizing images to better distinguish two domain samples.…”

Section: Related Workmentioning

confidence: 99%

Dual Progressive Prototype Network for Generalized Zero-Shot Learning

Wang

Min

Chen

et al. 2021

Preprint

View full text Add to dashboard Cite

Generalized Zero-Shot Learning (GZSL) aims to recognize new categories with auxiliary semantic information, e.g., category attributes. In this paper, we handle the critical issue of domain shift problem, i.e., confusion between seen and unseen categories, by progressively improving cross-domain transferability and category discriminability of visual representations. Our approach, named Dual Progressive Prototype Network (DPPN), constructs two types of prototypes that record prototypical visual patterns for attributes and categories, respectively. With attribute prototypes, DPPN alternately searches attribute-related local regions and updates corresponding attribute prototypes to progressively explore accurate attribute-region correspondence. This enables DPPN to produce visual representations with accurate attribute localization ability, which benefits the semantic-visual alignment and representation transferability. Besides, along with progressive attribute localization, DPPN further projects category prototypes into multiple spaces to progressively repel visual representations from different categories, which boosts category discriminability. Both attribute and category prototypes are collaboratively learned in a unified framework, which makes visual representations of DPPN transferable and distinctive. Experiments on four benchmarks prove that DPPN effectively alleviates the domain shift problem in GZSL.

show abstract

A Boundary Based Out-of-Distribution Classifier for Generalized Zero-Shot Learning

Cited by 49 publications

References 23 publications

Few-Shot Adaptive Detection of Objects of Concern Using Generative Models with Negative Retraining

Few-Shot Adaptive Detection of Objects of Concern Using Generative Models with Negative Retraining

Dual Prototype Relaxation for Generalized Zero Shot Learning

Dual Progressive Prototype Network for Generalized Zero-Shot Learning

Contact Info

Product

Resources

About