Escort: a system for rapid and long‐term tracking of multiple insect objects

All authors have reviewed and approved the letter, confirming their agreement with its contents. There are no conflicts of interest, including any specific financial interests, relationships or affiliations that could influence the subject matter of the letter. Fig. S1 The process for generating a list of unique, usable EStags. Fig. S2 Heat map depicting the self-optimized parameter tuning of adaptive thresholding using Gaussian kernel. Fig. S3 Three additional scenes for dataset preparation for YOLO 5 modelling training. Fig. S4 The architectural framework of the YOLO 5 model. Fig. S5 The distribution of the width and height of detection boxes in datasets. Fig. S6 The architectural framework of the YOLO 5 slim model. Table S1 Platform parameters for model training. Table S2 Performance of different YOLO models on test set. Table S3 Performance of EStag and Escort on videos captured in scenes 1, 2 and 3. Table S4 Platform parameters for tracking new datasets. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.