Reproduce “Fast Online Object Tracking and Segmentation: A Unifying Approach” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Qiang Wang, Li Zhang, Luca Bertinetto, Weiming Hu, Philip H.S. Torr
SiamMask extends an offline-trained Siamese tracker to predict object masks alongside tracking boxes, using pixel labels during training but only an initial bounding box at inference. It runs online without test-time updates. On VOT-2018, it achieves EAO 0.380 at 55 fps, versus 0.326 for DaSiamRPN; this supports real-time tracking, not fastest-tracker status.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.