ABSTRACT
For underwater videos, the performance of object tracking is greatly affected by illumination changes, background disturbances and occlusion. Hence, there is a need to have a robust function that computes image similarity, to accurately track the moving object. In this work, a hybrid model that incorporates the Kalman Filter, a Siamese neural network and a miniature neural network has been developed for object tracking. It was observed that the usage of the Siamese network to compute image similarity significantly improved the robustness of the tracker. Although the model was developed for underwater videos, it was found that it performs well for both underwater and human surveillance videos. A metric has been defined for analyzing detections-to-tracks mapping accuracy. Tracking results have been analyzed using Multiple Object Tracking Accuracy (MOTA) and Multiple Object Tracking Precision (MOTP)metrics.
- Yilmaz, Alper and Javed, Omar and Shah, Mubarak, "Object Tracking: A Survey", ACM Comput. Surv., Vol. 38, No. 4, 2006 Google ScholarDigital Library
- Yu Xiang, Alexandre Alahi and Silvio Savarese, "Learning to Track: Online Multi-Object Tracking by Decision Making", IEEE International Conference on Computer Vision, 2015, pp. 4705--4713 Google ScholarDigital Library
- Takala, Valtteri, and MattiPietikainen, "Multi object tracking using color, texture and motion". IEEE Conference on Computer Vision and Pattern Recognition. IEEE, 2007.Google ScholarCross Ref
- Han, Mei, et al. "A detection-based multiple object tracking method". Image Processing, 2004. ICIP'04. 2004 International Conference on. Vol. 5. IEEE, 2004.Google ScholarCross Ref
- Ondruska, Peter, and Ingmar Posner. "Deep tracking: Seeing beyond seeing using recurrent neural networks". arXiv preprint arXiv:1602.00991, 2016. Google ScholarDigital Library
- Huang, Phoenix X., Bastiaan J. Boom, and Robert B. Fisher, "Underwater live fish recognition using a balance-guaranteed optimized tree." Asian Conference on Computer Vision. Springer Berlin Heidelberg, 2012. Google ScholarDigital Library
- Fish-For-Knowledge @http://groups.inf.ed.ac.uk/f4k/Google Scholar
- Hemavathy R, Rashmi G N, Pooja Varambally, and Shobha G, "Segmenting of Object in postprocessing stage in Underwater Video Sequences Under Dynamic condition" in Proc IEEE Int. Advance Computing Conference(IACC), Jun2015, PgNo. 111--115Google Scholar
- Milan, Anton, et al., "MOT16: A Benchmark for Multi Object Tracking", arXiv preprint:1603.00831 (2016).Google Scholar
- Chopra, Sumit, Raia Hadsell, and Yann LeCun, "Learning a similarity metric discriminatively, with application to face verification", 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05). Vol. 1. IEEE, 2005. Google ScholarDigital Library
- Jia, Yangqing, et al. "Caffe: Convolutional architecture for fast feature embedding" Proceedings of the 22nd ACM international conference on Multimedia, 2014. Google ScholarDigital Library
- Yamaguchi, Kota https://github.com/kyamagu/mexopencvGoogle Scholar
- Hemavathy R, Rashmi G N, Shobha G, "MovingObject Segmentation from Underwater Videos Using Adaptive Collective Background Learning Approach", International Journal of Applied Engineering Research, Vol 11, No.9, pp. 6651--6654.Google Scholar
- Vondrick, Carl, Donald Patterson, and Deva Ramanan, "Efficiently scaling up crowdsourced video annotation", International Journal of Computer Vision 101.1, 2013, pp. 184--204. Google ScholarDigital Library
Recommendations
Partial tracking method based on siamese network
AbstractRobust object tracking is still a challenging task in the field of computer vision and has application value in many fields such as automatic driving, human–computer interaction and robot visual navigation. More and more researchers are devoted to ...
Template Attentional Siamese Network for Object Tracking
ICVIP '18: Proceedings of the 2018 2nd International Conference on Video and Image ProcessingRecent years, visual object tracking has attracted more and more attention as a fundamental topic. Many deep based trackers, especially Siamese Network based trackers, have achieved state-of-the-art performance on multiple benchmarks. However, most of ...
Occlusion aware underwater object tracking using hybrid adaptive deep SORT -YOLOv3 approach
AbstractUnderwater object tracking and recognition are challenging due to the distinctive characteristics of underwater environments. The water medium exhibits diffraction and scattering of light when it travels deep in the water. This results in unclear, ...
Comments