Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
139 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

SS-SFR: Synthetic Scenes Spatial Frequency Response on Virtual KITTI and Degraded Automotive Simulations for Object Detection (2407.15646v2)

Published 22 Jul 2024 in cs.CV

Abstract: Automotive simulation can potentially compensate for a lack of training data in computer vision applications. However, there has been little to no image quality evaluation of automotive simulation and the impact of optical degradations on simulation is little explored. In this work, we investigate Virtual KITTI and the impact of applying variations of Gaussian blur on image sharpness. Furthermore, we consider object detection, a common computer vision application on three different state-of-the-art models, thus allowing us to characterize the relationship between object detection and sharpness. It was found that while image sharpness (MTF50) degrades from an average of 0.245cy/px to approximately 0.119cy/px; object detection performance stays largely robust within 0.58\%(Faster RCNN), 1.45\%(YOLOF) and 1.93\%(DETR) across all respective held-out test sets.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (28)
  1. (2020). Zalazone. https://github.com/BMEAutomatedDrive/ZalaZONE-automotive-proving-ground-virtual-simulation-models. Accessed: 2024-04-22.
  2. (2024). Gaussianblur(). https://docs.opencv.org/3.4/d4/d86/group__imgproc__filter.html. Accessed: 2024-07-09.
  3. (2024). Mtf curves and image appearance. https://www.imatest.com/support/docs/23-1/mtf_appearance/. Accessed: 2024-05-02.
  4. Updated camera spatial frequency response for iso 12233. Electronic Imaging, 34:1–6.
  5. Virtual kitti 2. arXiv preprint arXiv:2001.10773.
  6. End-to-end object detection with transformers. In ECCV.
  7. Modeling camera effects to improve visual learning from synthetic data. In Computer Vision–ECCV 2018 Workshops: Munich, Germany, September 8-14, 2018, Proceedings, Part I 15, pages 505–520. Springer.
  8. MMDetection: Open mmlab detection toolbox and benchmark. arXiv preprint arXiv:1906.07155.
  9. You only look one-level feature. In IEEE Conference on Computer Vision and Pattern Recognition.
  10. cocodataset.org ((n.d.))). Detection evaluation. https://cocodataset.org/#detection-eval. Accessed: 2024-07-22.
  11. Slanted edge method for mtf measurements in the infrared. Infrared Physics & Technology, 118:103877.
  12. Virtual worlds as proxy for multi-object tracking analysis. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4340–4349.
  13. Vision meets robotics: The kitti dataset. The International Journal of Robotics Research, 32(11):1231–1237.
  14. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770–778.
  15. Benchmarking neural network robustness to common corruptions and perturbations. arXiv preprint arXiv:1903.12261.
  16. Institute, B. S. (2023). BS ISO 12233:2023 Photography - Electronic still picture imaging - Resolution and spatial frequency responses. BSI Standards Publication.
  17. Surround-view fisheye optics in computer vision and simulation: Survey and challenges. IEEE Transactions on Intelligent Transportation Systems, pages 1–22.
  18. Measuring natural scenes sfr of automotive fisheye cameras. Electronic Imaging, 36:1–6.
  19. Simulating tests to test simulation. Electronic Imaging, 32:1–8.
  20. Asam opendrive. https://www.asam.net/standards/detail/opendrive/. Accessed: 2024-04-22.
  21. Faster r-cnn: Towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence.
  22. Camera vignetting model and its effects on deep neural networks for object detection. In 2019 IEEE International Conference on Connected Vehicles and Expo (ICCVE), pages 1–5. IEEE.
  23. Szeliski, R. (2011). Computer vision algorithms and applications.
  24. van Zwanenberg, O. (2022). Camera spatial frequency response derived from pictorial natural scenes. PhD thesis, University of Westminster.
  25. Natural scene derived camera edge spatial frequency response for autonomous vision systems. In IS&T/IoP London Imaging Meeting.
  26. A tool for deriving camera spatial frequency response from natural scenes (ns-sfr). Electronic Imaging, 35:1–6.
  27. Bdd100k: A diverse driving dataset for heterogeneous multitask learning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 2636–2645.
  28. Zernike, F. (1934). Beugungstheorie des schneidenver-fahrens und seiner verbesserten form, der phasenkontrastmethode. physica, 1(7-12):689–704.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now