Flexible dual-branch Siamese network: learning location quality estimation and regression distribution for visual tracking

Shuo Hu, Sien Zhou, Jinbo Lu, Hui Yu

Research output: Contribution to journalArticlepeer-review

31 Downloads (Pure)


Anchor-free based trackers introduce an extra branch in addition to classification and regression branches in the network to achieve comparable performance with anchor-based trackers. This extra branch is usually trained independently in the training phase and is used in combination with other branches in the inference phase. However, this can increase the inconsistency between the inference phase and the training phase, potentially degrading the tracking performance. To address this problem, we propose a new Siamese network-based object tracking framework that eliminates this inconsistency by unifying classification and additional branch tasks to achieve learning location quality estimation. Furthermore, regression tasks for bounding boxes are widely formulated based on Dirac δ distribution. Though this assumption works well for many scenarios, it restricts the prediction of regression branches. To overcome this restriction, we propose discretizing the continuous offset of the regression branch into multiple offset predictions, which enables the network to learn more flexible distributions automatically. Meanwhile, the discrete distribution prediction of regression branches is utilized to further guide the classification of the trackers. Extensive experiments on the widely accepted benchmarks demonstrate the effectiveness and efficiency of the proposed model.
Original languageEnglish
Pages (from-to)1-9
Number of pages9
JournalIEEE Transactions on Computational Social Systems
Early online date31 Jan 2023
Publication statusEarly online - 31 Jan 2023


  • Anchor-free
  • distributed guidance
  • estimation
  • feature extraction
  • IOU-aware classification
  • object tracking
  • Siamese network
  • target tracking
  • task analysis
  • training
  • visualization

Cite this