Research Article

Hybrid Approach for Shelf Monitoring and Planogram Compliance (Hyb-SMPC) in Retails Using Deep Learning and Computer Vision

Table 1

Comparison of proposed approach with the previous approaches.

Reference no.YearObject detectionPlanogram complianceMethods
Traditional methodDeep learning
Two-stageOne-stage

[4]2015SURF
[16]2015SURF
[18]2015SVM
[19]2015SURF + color histogram
[22]2016Recurring pattern detection
[38]2016CNN
[36]2017VGG-F
[17]2018SIFT
[21]2018VGG-16 with recurring features and attention maps
[37]2018GoogLeNet
[35]2019CIFAR-10, CaffeNet
[43]2020RetinaNet, YOLO V3, YOLO V4
[44]2020VGG-16, Tiny YOLO V2
Proposed methodYOLO V4, YOLO V5, YOLOR