A Siamese Network for real-time object tracking on CPU[Formula presented]

Daitao Xing, Nikolaos Evangeliou, Athanasios Tsoukalas, Anthony Tzes

Research output: Contribution to journalArticlepeer-review


Visual object tracking methods depend upon deep networks that can hardly meet real-time processing requirements on mobile platforms with limited computing resources. In this work, we propose a real-time object tracking framework by enhancing a lightweight feature pyramid network with Transformer architecture to construct a robust target-specific appearance model efficiently. We further introduce the pooling attention module to avoid the computation and memory intensity while fusing pyramid features with the Transformer. The optimized tracker operates over 45 Hz on a single CPU, allowing researchers to deploy it on any mobile device with limited power resources.

Original languageEnglish (US)
Article number100266
JournalSoftware Impacts
StatePublished - May 2022


  • Mobile devices
  • Object tracking
  • Reduced complexity
  • Siamese Network

ASJC Scopus subject areas

  • Software


Dive into the research topics of 'A Siamese Network for real-time object tracking on CPU[Formula presented]'. Together they form a unique fingerprint.

Cite this