A Co-Training Semi-Supervised Framework Using Faster R-CNN and YOLO Networks for Object Detection in Densely Packed Retail Images

2025-09-11

AI Summary

To address degraded object detection performance in retail dense-stacking scenarios—caused by label scarcity, severe occlusion, and dynamic layout changes—this paper proposes a semi-supervised co-training framework. The method integrates Faster R-CNN (with ResNet backbone) and YOLO (with Darknet backbone), introducing a novel dual-model pseudo-label mutual-feedback mechanism. It further incorporates a multimodal feature classifier composed of XGBoost, Random Forest, and SVM, and employs metaheuristic algorithms for hyperparameter optimization. Evaluated on the SKU-110K dataset, the framework significantly reduces reliance on labeled data while enhancing model generalization and robustness. Experimental results demonstrate high accuracy and strong industrial deployability in real-world retail applications, including automated inventory auditing, product monitoring, and cashier-less checkout systems.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

This study proposes a semi-supervised co-training framework for object detection in densely packed retail environments, where limited labeled data and complex conditions pose major challenges. The framework combines Faster R-CNN (utilizing a ResNet backbone) for precise localization with YOLO (employing a Darknet backbone) for global context, enabling mutual pseudo-label exchange that improves accuracy in scenes with occlusion and overlapping objects. To strengthen classification, it employs an ensemble of XGBoost, Random Forest, and SVM, utilizing diverse feature representations for higher robustness. Hyperparameters are optimized using a metaheuristic-driven algorithm, enhancing precision and efficiency across models. By minimizing reliance on manual labeling, the approach reduces annotation costs and adapts effectively to frequent product and layout changes common in retail. Experiments on the SKU-110k dataset demonstrate strong performance, highlighting the scalability and practicality of the proposed framework for real-world retail applications such as automated inventory tracking, product monitoring, and checkout systems.

Problem

• Semi-supervised object detection in densely packed retail images
• Addressing limited labeled data and complex occlusion conditions
• Reducing manual annotation costs for retail product recognition

Innovation

• Semi-supervised co-training with Faster R-CNN and YOLO
• Ensemble classification using XGBoost, Random Forest, SVM
• Metaheuristic algorithm for hyperparameter optimization