Using Unlabeled Data for Increasing Low-Shot Classification Accuracy of Relevant and Open-Set Irrelevant Images

Using Unlabeled Data for Increasing Low-Shot Classification Accuracy of Relevant and Open-Set Irrelevant Images

Spiridon Kasapis

Contact

Geng Zang

Contact

Jonathon M. Smereka

Contact

Nickolas Vlahopoulos

Contact

University of Michigan

Using Unlabeled Data for Increasing Low-Shot Classification Accuracy of Relevant and Open-Set Irrelevant Images

Article Fingerprint

ReserarchID

G070O

Using Unlabeled Data for Increasing Low-Shot Classification Accuracy of Relevant and Open-Set Irrelevant Images Banner

AI TAKEAWAY

Connecting with the Eternal Ground

Abstract

In search, exploration, and reconnaissance tasks performed with autonomous ground vehicles, an image classification capability is needed for specifically identifying targeted objects (relevant classes) and at the same time recognize when a candidate image does not belong to anyone of the relevant classes (irrelevant images). In this paper, we present an open-set low-shot classifier that uses, during its training, a modest number (less than 40) of labeled images for each relevant class, and unlabeled irrelevant images that are randomly selected at each epoch of the training process. The new classifier is capable of identifying images from the relevant classes, determining when a candidate image is irrelevant, and it can further recognize categories of irrelevant images that were not included in the training (unseen). The proposed lowshot classifier can be attached as a top layer to any pre-trained feature extractor when constructing a Convolutional Neural Network.

References

36 Cites in Article

Reference Format

Hossein Azizpour (2015). From generic to specific deep representations for visual recognition.
Basha Sh Shabbeer (2021). AutoTune: Automatically Tuning Convolutional Neural Networks for Improved Transfer Learning.
Amir Beck,Marc Teboulle (2009). A Fast Iterative Shrinkage-Thresholding Algorithm for Linear Inverse Problems.
Abhijit Bendale,Terrance Boult (2016). Towards Open Set Deep Networks.
Naresh Vishnu,Boddeti,Kumar (2014). Maximum margin vector correlation filter.
Jia Deng,Wei Dong,Richard Socher,Li-Jia Li,Kai Li,Li Fei-Fei (2009). ImageNet: A large-scale hierarchical image database.
Raj Akshay,Manuel Dhamija,Terrancee Günther,Boult (2018). Reducing network agnostophobia.
Steven Tang,Joshua Peterson,Zachary Pardos (2017). Predictive Modelling of Student Behavior Using Granular Large-Scale Action Data.
Weifeng Ge,Yizhou Yu (2017). Borrowing Treasures from the Wealthy: Deep Transfer Learning through Selective Joint Fine-Tuning.
Zongyuan Ge,Sergey Demyanov,Rahil Garnavi (2017). Generative OpenMax for Multi-Class Open Set Classification.
Gregory Griffin,Alex Holub,Pietro Perona (2007). Caltech-256 object category dataset.
J David,Hand (2009). Measuring classifier performance: a coherent alternative to the area under the ROC curve.
J David,Robert Hand,Till (2001). A simple generalization of the area under the ROC curve for multiple class classification problems.
Sarah Harris,David Harris (2010). Architecture.
Kaiming He,Xiangyu Zhang,Shaoqing Ren,Jian Sun (2016). Deep Residual Learning for Image Recognition.
Kaiming He (2016). Identity mappings in deep residual networks.
Kaiming He,Georgia Gkioxari,Piotr Dollar,Ross Girshick (2017). Mask R-CNN.
Dan Hendrycks,Kevin Gimpel (2016). A baseline for detecting misclassified and out-of-distribution examples in neural networks.
Lei Huang (2018). Orthogonal weight normalization: Solution to optimization over multiple dependent stiefel manifolds in deep neural networks.
Walter Lalit P Jain,Terrance Scheirer,Boult (2014). Multi-class open set recognition using probability of inclusion.
Kui Jia (2017). Improving training of deep neural networks via singular value bounding.
Hong Kim,Nickolas Vlahopoulos (2012). A Multi-Level Optimization Algorithm and a Ship Design Application.
Jedrzej Kozerawski,Matthew Turk (2021). One-Class Meta-Learning: Towards Generalizable Few-Shot Open-Set Classification.
Alex Krizhevsky,Ilya Sutskever,Geoffrey Hinton (2012). ImageNet classification with deep convolutional neural networks.
Bo Liu,Hao Kang,Haoxiang Li,Gang Hua,Nuno Vasconcelos (2020). Few-Shot Open-Set Recognition Using Meta-Learning.
Jiajun Lu,Theerasit Issaranon,David Forsyth (2017). SafetyNet: Detecting and Rejecting Adversarial Examples Robustly.
Dj Peres,Cancelliere (2014). Derivation and evaluation of landslide-triggering thresholds by a Monte Carlo approach.
Federico Pernici,Federico Bartoli,Matteo Bruni,Alberto Del Bimbo (2018). Memory Based Online Learning of Deep Representations from Video Streams.
Iq Pham,M Polasek (2014). Algorithm for military object detection using image data.
J Walter,Lalit Scheirer,Terrance Jain,Boult (2014). Probability models for open set recognition.
J Walter,Scheirer (2012). Toward open set recognition.
Karen Simonyan,Andrew Zisserman (2014). Preprint repository arXiv achieves milestone million uploads.
Le Wang,Gang Hua,Rahul Sukthankar,Jianru Xue,Nanning Zheng (2014). Video Object Discovery and Co-segmentation with Extremely Weak Supervision.
Liangjiang Yu,Guoliang Fan,Jiulu Gong,Joseph Havlicek (2015). Joint Infrared Target Recognition and Segmentation Using a Shape Manifold-Aware Level Set.
Bolei Zhou,Agata Lapedriza,Aditya Khosla,Aude Oliva,Antonio Torralba (2017). Places: A 10 Million Image Database for Scene Recognition.
Wangmeng Zuo,Xiaohe Wu,Liang Lin,Lei Zhang,Ming-Hsuan Yang (2018). Learning Support Correlation Filters for Visual Tracking.

Download References

Funding

No external funding was declared for this work.

Conflict of Interest

The authors declare no conflict of interest.

Ethical Approval

No ethics committee approval was required for this article type.

Data Availability

Not applicable for this article.

How to Cite This Article

Spiridon Kasapis. 2026. \u201cUsing Unlabeled Data for Increasing Low-Shot Classification Accuracy of Relevant and Open-Set Irrelevant Images\u201d. Global Journal of Computer Science and Technology - D: Neural & AI GJCST-D Volume 22 (GJCST Volume 22 Issue D2): .

More Citation Formats

Select Citation Style:

Download Citation

Download Article

GJCST Volume 22 Issue D2
Pg. 11- 24

Explore Journals Explore Volume Read This Issue

Journal Specifications

Crossref Journal DOI 10.17406/gjcst

Print ISSN 0975-4350

e-ISSN 0975-4172

Keywords

Not Found

Classification

GJCST-D Classification: DDC Code: 025.0425 LCC Code: ZA3075

Submission ReceivedMarch 22, 2022
Peer Review Double Blind
Handling Editor
Accepted March 29, 2022
Published May 13, 2022

Version of record

v1.2

Issue date

May 26, 2022

Language

Experiance in AR

Explore published articles in an immersive Augmented Reality environment. Our platform converts research papers into interactive 3D books, allowing readers to view and interact with content using AR and VR compatible devices.

View in VR

Read in 3D

Your published article is automatically converted into a realistic 3D book. Flip through pages and read research papers in a more engaging and interactive format.

View in 3D

Article Matrices

Total Views: 2839

Total Downloads: 41

2026 Trends

Published Article

Our website is actively being updated, and changes may occur frequently. Please clear your browser cache if needed. For feedback or error reporting, please email [email protected]