High-level Performance Evaluation of Object Detection based on Massively Parallel Focal-plane Acceleration Requiring Minimum Pixel Area Overhead

Eloy Parra-Barrero, Jorge Fernández-Berni, Fernanda D. V. R. Oliveira, Ricardo Carmona-Galán, Ángel Rodríguez-Vázquez

Abstract

Smart CMOS image sensors can leverage the inherent data-level parallelism and regular computational flow of early vision by incorporating elementary processors at pixel level. However, it comes at the cost of extra area having a strong impact on the sensor sensitivity, resolution and image quality. In this scenario, the fundamental challenge is to devise new strategies capable of boosting the performance of the targeted vision pipeline while minimally affecting the sensing function itself. Such strategies must also feature enough flexibility to accommodate particular application requirements. From these high-level specifications, we propose a focal-plane processing architecture tailored to speed up object detection via the Viola-Jones algorithm. This architecture is supported by only two extra transistors per pixel and simple peripheral digital circuitry that jointly make up a massively parallel reconfigurable processing lattice. A performance evaluation of the proposed scheme in terms of accuracy and acceleration for face detection is reported.

References

  1. Abramson, Y., Steux, B., and Ghorayeb, H. (2007). Yet even faster (YEF) real-time object detection. Int. J. of Intelligent Systems Technologies and Applications, 2(2-3):102-112.
  2. Bradski, G. (2000). The OpenCV library. Dr. Dobbs Journal of Software Tools.
  3. Camilli, M. and Kleihorst, R. (2011). Demo: Mouse sensor networks, the smart camera. In 5th ACM/IEEE Int. C. on Distributed Smart Cameras, Ghent, Belgium.
  4. Fernández-Berni, J., Carmona-Galán, R., and CarranzaGonzález, L. (2011). FLIP-Q: A QCIF resolution focal-plane array for low-power image processing. IEEE Int. Journal of Solid-State Circuits, 46(3):669- 680.
  5. Fowler, B. (2015). Solid-state image sensors. In Kriss, M., editor, Handbook of Digital Imaging. John Wiley & Sons, Ltd.
  6. Jia, H., Zhang, Y., Wang, W., and Xu, J. (2012). Accelerating viola-jones face detection algorithm on GPUs. In IEEE Int. Conf. on Embedded Software and Systems, pages 396-403.
  7. Klette, R. (2014). Concise Computer Vision. Springer.
  8. Ohta, J. (2007). Smart CMOS Image Sensors and Applications. CRC Press.
  9. Ouyang, P., Yin, P., Yin, S., Zhang, Y., Liu, L., and Wei, S. (2015). A fast integral image computing hardware architecture with high power and area efficiency.IEEE Transactions on Circuits and Systems II, 62(1):75-79.
  10. Viola, P. and Jones, M. (2004). Robust real-time face detection. Int. J. of Computer Vision, 57(2):137-154.
  11. Weber, M. (1999). Caltech frontal face dataset. http://www.vision.caltech.edu/htmlfiles/archive.html.
  12. Zarándy, A., editor (2011). Focal-plane Sensor-Processor Chips. Springer.
Download


Paper Citation


in Harvard Style

Parra-Barrero E., Fernández-Berni J., Oliveira F., Carmona-Galán R. and Rodríguez-Vázquez Á. (2016). High-level Performance Evaluation of Object Detection based on Massively Parallel Focal-plane Acceleration Requiring Minimum Pixel Area Overhead . In Proceedings of the 11th Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 3: VISAPP, (VISIGRAPP 2016) ISBN 978-989-758-175-5, pages 79-85. DOI: 10.5220/0005651200790085


in Bibtex Style

@conference{visapp16,
author={Eloy Parra-Barrero and Jorge Fernández-Berni and Fernanda D. V. R. Oliveira and Ricardo Carmona-Galán and Ángel Rodríguez-Vázquez},
title={High-level Performance Evaluation of Object Detection based on Massively Parallel Focal-plane Acceleration Requiring Minimum Pixel Area Overhead},
booktitle={Proceedings of the 11th Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 3: VISAPP, (VISIGRAPP 2016)},
year={2016},
pages={79-85},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0005651200790085},
isbn={978-989-758-175-5},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 11th Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 3: VISAPP, (VISIGRAPP 2016)
TI - High-level Performance Evaluation of Object Detection based on Massively Parallel Focal-plane Acceleration Requiring Minimum Pixel Area Overhead
SN - 978-989-758-175-5
AU - Parra-Barrero E.
AU - Fernández-Berni J.
AU - Oliveira F.
AU - Carmona-Galán R.
AU - Rodríguez-Vázquez Á.
PY - 2016
SP - 79
EP - 85
DO - 10.5220/0005651200790085