liu.seSearch for publications in DiVA
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Generative inter-class transformations for imbalanced data weather classification
Linköping University, Department of Science and Technology, Media and Information Technology. Linköping University, Faculty of Science & Engineering.ORCID iD: 0000-0003-0298-937X
Linköping University.
Linköping University.
Linköping University, Department of Science and Technology, Media and Information Technology. Linköping University, Faculty of Science & Engineering.ORCID iD: 0000-0002-9217-9997
Show others and affiliations
2021 (English)In: London Imaging Meeting, E-ISSN 2694-118X, Vol. 2021, p. 16-20Article in journal (Refereed) Published
Abstract [en]

This paper presents an evaluation of how data augmentation and inter-class transformations can be used to synthesize training data in low-data scenarios for single-image weather classification. In such scenarios, augmentations is a critical component, but there is a limit to how much improvements can be gained using classical augmentation strategies. Generative adversarial networks (GAN) have been demonstrated to generate impressive results, and have also been successful as a tool for data augmentation, but mostly for images of limited diversity, such as in medical applications. We investigate the possibilities in using generative augmentations for balancing a small weather classification dataset, where one class has a reduced number of images. We compare intra-class augmentations by means of classical transformations as well as noise-to-image GANs, to interclass augmentations where images from another class are transformed to the underrepresented class. The results show that it is possible to take advantage of GANs for inter-class augmentations to balance a small dataset for weather classification. This opens up for future work on GAN-based augmentations in scenarios where data is both diverse and scarce.

Place, publisher, year, edition, pages
Springfield, USA: Society for Imaging Science and Technology , 2021. Vol. 2021, p. 16-20
National Category
Computer graphics and computer vision
Identifiers
URN: urn:nbn:se:liu:diva-182334DOI: 10.2352/issn.2694-118X.2021.LIM-16OAI: oai:DiVA.org:liu-182334DiVA, id: diva2:1628943
Note

Funding: This project was funded by Knut and Alice Wallenberg Foundation, Wallenberg Autonomous Systems and Software Program, the strategic research environment ELLIIT, and ‘AI for Climate Adaptation’ through VINNOVA grant 2020-03388.

Available from: 2022-01-17 Created: 2022-01-17 Last updated: 2025-02-07Bibliographically approved
In thesis
1. Synthetic data for visual machine learning: A data-centric approach
Open this publication in new window or tab >>Synthetic data for visual machine learning: A data-centric approach
2022 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Deep learning allows computers to learn from observations, or else training data. Successful application development requires skills in neural network design, adequate computational resources, and a training data distribution that covers the application do-main. We are currently witnessing an artificial intelligence (AI) outbreak with enough computational power to train very deep networks and build models that achieve similar or better than human performance. The crucial factor for the algorithms to succeed has proven to be the training data fed to the learning process. Too little or low quality or out-of-the-target distribution data will lead to poorly performing models no matter the capacity and the data regularization methods.

This thesis is a data-centric approach to AI and presents a set of contributions related to synthesizing images for training supervised visual machine learning. It is motivated by the profound potential of synthetic data in cases of low availability of captured data, expensive acquisition and annotation, and privacy and ethical issues. The presented work aims to generate images similar to samples drawn from the target distribution and evaluate the generated data as the sole training data source and in conjunction with captured imagery. For this, two synthesis methods are explored: computer graphics and generative modeling. Computer graphics-based generation methods and synthetic datasets for computer vision tasks are thoroughly reviewed. In the same context, a system employing procedural modeling and physically-based rendering is introduced for data generation for urban scene understanding. The scheme is flexible, easily scalable, and produces complex and diverse images with pixel-perfect annotations at no cost. Generative Adversarial Networks (GANs) are also used to generate images for small data scenarios augmentation. The strategy advances the model’s performance and robustness. Finally, ensembles of independently trained GANs investigate ways to improve images’ diversity and create synthetic data to serve as the only training source.

The application areas of the presented contributions relate to two image modalities, natural and histopathology images, to cover different aspects in the generation methods and the tasks’ characteristics and requirements. There are showcased synthesized examples of natural images for automotive applications and weather classification, and histopathology images for breast cancer and colon adenocarcinoma metastasis detection. This thesis, as a whole, promotes data-centric supervised deep learning development by highlighting the potential of synthetic data as a training data resource. It emphasizes the control over the formation process, the ability of multi-modality formats, and the automatic generation of annotations.

Place, publisher, year, edition, pages
Linköping: Linköping University Electronic Press, 2022. p. 115
Series
Linköping Studies in Science and Technology. Dissertations, ISSN 0345-7524 ; 2202
Keywords
Training data, Synthetic images, Computer graphics, Generative modeling, Natural images, Histopathology, Digital pathology, Machine learning, Deep learning
National Category
Medical Imaging
Identifiers
urn:nbn:se:liu:diva-182336 (URN)10.3384/9789179291754 (DOI)9789179291747 (ISBN)9789179291754 (ISBN)
Public defence
2022-02-14, Domteatern, Visualiseringscenter C, Kungsgatan 54, Norrköping, 09:15 (English)
Opponent
Supervisors
Note

ISBN for PDF has been added in the PDF-version.

Available from: 2022-01-17 Created: 2022-01-17 Last updated: 2025-02-09Bibliographically approved

Open Access in DiVA

fulltext(1459 kB)207 downloads
File information
File name FULLTEXT01.pdfFile size 1459 kBChecksum SHA-512
9b1353f503fbab31aaacf9fe7d29c3ac886929c94980f9778a83168c728ed70eebc0310146d08efb44f2f84dcabfa20615f29a453cdd02839880ba16ac49f3e9
Type fulltextMimetype application/pdf

Other links

Publisher's full text

Authority records

Tsirikoglou, ApostoliaEilertsen, GabrielUnger, Jonas

Search in DiVA

By author/editor
Tsirikoglou, ApostoliaGladh, MarcusSahlin, DanielEilertsen, GabrielUnger, Jonas
By organisation
Media and Information TechnologyFaculty of Science & EngineeringLinköping University
Computer graphics and computer vision

Search outside of DiVA

GoogleGoogle Scholar
Total: 207 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 319 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf