Learning Rotation Sensitive Neural Network for Deformed Objects' Detection in Fisheye Images

Research output: Contributions to collected editions/worksArticle in conference proceedingsResearchpeer-review

Authors

Object detection plays a significant role in an intelligent system equipped with a fisheye camera. The fisheye image captures a wide field-of-view but deforms in the radial direction. The deformation changes the relative angle between the edge of objects and the image. Therefore, a horizontal bounding box cannot perform an accurate description of an object's location and dimension in advanced neural network training. In this paper, we build a rotation sensitive neural network targeting to realize one-stage regression on the fisheye image detection. The oriented bounding box is applied in the object's description and detection. To evaluate our proposed method, we develop a new labelled fisheye image dataset that contains two categories. The network model training takes around 3 hours and achieves 100% precious by the test set.

Original languageEnglish
Title of host publication2019 4th International Conference on Robotics and Automation Engineering (ICRAE 2019) : November 22-24, 2019, Singapore
Number of pages5
Place of PublicationPiscataway
PublisherIEEE - Institute of Electrical and Electronics Engineers Inc.
Publication date11.2019
Pages125-129
Article number9043800
ISBN (print)978-1-7281-4741-3
ISBN (electronic)978-1-7281-4740-6, 978-1-7281-4739-0
DOIs
Publication statusPublished - 11.2019
Event4th International Conference on Robotics and Automation Engineering, ICRAE 2019 - Nanyang Technological University, Singapore, Singapore, Singapore
Duration: 22.11.201924.11.2019
Conference number: 4
https://www.ieee-ras.org/component/rseventspro/event/1695-icrae-2019-international-conference-on-robotics-and-automation-engineering

    Research areas

  • Fisheye image, Image distortion, Orientated Box, YOLO detector
  • Engineering

Recently viewed

Publications

  1. High resolution measurement of physical variables change for INS
  2. Simulation of stresses during casting of binary magnesium-aluminum alloys
  3. Social group membership does not modulate automatic imitation in a contrastive multi-agent paradigm
  4. Organizing Half-Things: Knowing, Theorizing and Studying Atmospheres
  5. Using a decoupling technique to identify the magnetic flux in a permanent magnet synchronous motor
  6. The State of Multimedia Mass-Balance Modeling in Environmental science and decision-making
  7. Properties of some overlapping self-similar and some self-affine measures
  8. Microstructure and mechanical properties of large-scale Mg-Gd-Y-Zn-Mn alloys prepared through semi-continuous casting
  9. Connecting curricula and competence through student learning journeys
  10. Spielt Charlie Parker in den Wind oder mit ihm?
  11. Estimation of phosphorus export from a Mediterranean agricultural catchment with scarce data
  12. Shape-shifting
  13. Factorial Validity of the Anxiety Questionnaire for Students (AFS)
  14. Selbstöffnung und interpersonaler Input in der psychodynamischen Gruppenpsychotherapie – Eine explorative Studie
  15. Corrosion behaviour of magnesium alloys with RE additions in sodium chloride solutions
  16. Texturig Space
  17. The new US horizontal merger guidelines
  18. Diversity matters
  19. An environmentally friendly procedure to obtain flavonoids from Brazilian citrus waste
  20. Keeping in touch
  21. (Not) Thinking about you
  22. Which clicks lead to conversions?
  23. "(Nicht) alles dreht sich um Harry ..." oder was fasziniert Kinder, Jugendliche und Erwachsene an der Welt des Harry Potter?
  24. The Great Pacific Garbage Catch
  25. Nitrogen cycling and storage in Gagea spathacea (Liliaceae)
  26. 'So tun als ob' 2.0