Improved YOLOv3-tiny for Silhouette Detection Using Regularisation Techniques

[摘要] Although recent advances in Deep Learning (DL) algorithms have been developed in many Computer Vision (CV)tasks with a high accuracy level, detecting humans in video streams is still a challenging problem. Several studies have,therefore, focused on the regularisation techniques to prevent the overfitting problem which is one of the most fundamentalissues in the Machine Learning (ML) area. Likewise, this paper thoroughly examines these techniques, suggesting an improvedyou Only Look Once (YOLO) v3-tiny based on a modified neural network and an adjusted hyperparameters file configuration.The obtained experimental results, which are validated on two experimental tests, show that the proposed method is moreeffective than the YOLOv3-tiny predecessor model. The first test which includes only the data augmentation techniquesindicates that the proposed approach reaches higher accuracy rates than the original YOLOv3-tiny model. Indeed, VisualObject Classes (VOC) test dataset accuracy rate increases by 32.54 % compared to the initial model. The second test whichcombines the three tasks reveals that the adopted combined method wins a gain over the existing model. For instance, thelabelled crowd_human test dataset accuracy percentage rises by 22.7 % compared to the data augmentation model.

[学科分类] 计算机科学（综合）

[关键词] Silhouette/person detection;GPU;loss function;convolutional neural network;YOLOv3-tiny