Embedded Visual Object Tracking System Based on CSRT Tracker

Ilkin S., Kaya Gülağız F., Akcakaya M., Şahin S.

2022 International Conference on Electronics, Information, and Communication, ICEIC 2022, Jeju, Güney Kore, 6 - 09 Şubat 2022 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/iceic54506.2022.9748840
  • Basıldığı Şehir: Jeju
  • Basıldığı Ülke: Güney Kore
  • Anahtar Kelimeler: object tracking, embedded systems, IoT, motion tracking, CSRT tracker
  • Kocaeli Üniversitesi Adresli: Evet

Özet

© 2022 IEEE.Nowadays, the importance given to object detection and tracking studies is increasing. With the developing technol-ogy, the IoT systems that have gained an irresistible place in our lives are started to be used in these studies too. In this paper, we aim to measure the performance analysis of object tracking process on the videos via embedded systems. We used Jetson NANO and Raspberry Pi development boards for this purpose. The CSRT tracker was used for object tracking. The system was tested by two video datasets which are OTH-15 and LaSOT. Experimental results shows that Jetson NANO development kit is promising for object tracking on the videos with the average accuracy rate of 0.57 for OTH-15 dataset and 0.30 for LaSOT dataset.