SINGLE STAGE DEEP TRANSFER LEARNING MODEL FOR APPAREL DETECTION AND CLASSIFICATION FOR E-COMMERCE
Ssvr Kumar Addagarla, Anthoniraj Amalanathan · International Journal of Electronic Commerce Studies · 2021
Although many computer vision based object detection techniques are evolved in the past decade but suffers from inconsistent detection accuracy especially for multi-class classification problems. In this paper proposed an approach using Single Stage Deep Transfer Learning model (SS-DTLM) for multi-class apparel detection using customized YoloV3 algorithm by adapting 3-level Spatial pyramid pooling (SPP), a multi scale image feature extractor for faster and reasonable apparel detection and classification. This approach produced a reasonable Mean Average Precision (mAP), reliable object detection and classification. Our model trained and tested on Open Images Dataset (OIDV4) with 6 object classes and Custom built Apparel Dataset with 5 object classes of apparels. Finally Experimental Results are compared with base line Yolov3 and Yolov3-Tiny algorithms. Further this paper also emphasized various color spaces of the detected image using SS-DTLM by applying K-Means clustering algorithm for further analysis.