A neuro-inspired visual tracking method based on programmable system-on-chip platform
- 249 Downloads
Using programmable system-on-chip to implement computer vision functions poses many challenges due to highly constrained resources in cost, size and power consumption. In this work, we propose a new neuro-inspired image processing model and implemented it on a system-on-chip Xilinx Z702c board. With the attractor neural network model to store the object’s contour information, we eliminate the computationally expensive steps in the curve evolution re-initialisation at every new iteration or frame. Our experimental results demonstrate that this integrated approach achieves accurate and robust object tracking, when they are partially or completely occluded in the scenes. Importantly, the system is able to process 640 by 480 videos in real-time stream with 30 frames per second using only one low-power Xilinx Zynq-7000 system-on-chip platform. This proof-of-concept work has demonstrated the advantage of incorporating neuro-inspired features in solving image processing problems during occlusion.
KeywordsVisual object tracking Mean-shift Level set Attractor neural network model Occlusion System-on-chip
SY was supported by the Early Research Scheme Reward from University of Wolverhampton and National High Technology Research and Development Program from China. KFW-L and SY were supported by ASUR (1014-C4-Ph1-071).
- 2.Nikitakis A, Papaioannou S, Papaefstathiou I (2013) A novel low-power embedded object recognition system working at multi-frames per second. ACM Trans Embed Comput Syst (TECS) 12(33):39–58Google Scholar
- 4.Jin J, Lee S, Jeon B, Nguyen TT, Jeon JW (2013) Real-time multiple object centroid tracking for gesture recognition based on FPGA. In: Proceedings of the 7th international conference on ubiquitous information management and communication, article 80Google Scholar
- 5.Nguyen HT, Smeulders A (2004) Tracking aspects of the foreground against the background. In Computer Vision-ECCV 2004. Springer, Berlin, pp 446–456Google Scholar
- 6.Lee BY, Liew LH, Cheah WS, Wang YC (2014) Occlusion handling in videos object tracking: a survey. In IOP conference series: earth and environmental science, vol 18(1). IOP Publishing, p 012020Google Scholar
- 8.Yantis S, Johnson DN (1990) Mechanisms of attentional priority. J Exp Psychol Hum Percept Perform 16(4):812–825Google Scholar
- 13.Dalal N, Triggs B (2005) Histograms of oriented gradients for human detection. In: Computer vision and pattern recognition, 2005. CVPR 2005. IEEE computer society conference on IEEE, vol 1, pp 886–893Google Scholar
- 15.Han M, Xu W, Gong Y (2004) An algorithm for multiple object trajectory tracking. Proc IEEE Comput Soc Conf 1:864–871Google Scholar
- 16.Chan TE, Vese LA (2001) A level set algorithm for minimizing the Mumford–Shah functional in image processing. In: Variational and level set methods in computer vision, 2001. Proceedings of IEEE Workshop on IEEE, pp 161–168Google Scholar
- 22.Xilinx (2014) Xilinx Xpower analyzer. www.xilinx.com/products
- 24.Dollár P, Wojek C, Schiele B, Perona P (2009) Pedestrian detection: a benchmark. In: Computer vision and pattern recognition, 2009. CVPR 2009. IEEE conference on IEEE, pp 304–311Google Scholar