Scholarlink

This study investigates the application of Federated Learning alongside You Only Look Once (YOLO) models for identifying and classifying solid food items to estimate calorie content. By leveraging existing technology, specifically the YOLO model, the research aims to detect countable solid foods and calculate their caloric values using a specialized dataset. Different YOLO configurations are evaluated to determine their accuracy in detecting and classifying food items. Federated learning facilitates decentralized training on local datasets, transmitting only model updates to a central server, thus preserving data privacy and meeting regulatory requirements. This method fosters collaborative learning across distributed devices.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Enhanced Calorie Estimation of Solid Foods Using Federated Learning and YOLO Models: A Distributed Approach for Collaborative Caloric Data Analysis

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Md. Ashiq Ul Islam Sajid,
  • Abdullah All Mubin,
  • Mohammed Aftahi Islam,
  • Ripa Sarkar,
  • Sumaya Binte Zilani Choya,
  • Md. Tamim Hasan

摘要

This study investigates the application of Federated Learning alongside You Only Look Once (YOLO) models for identifying and classifying solid food items to estimate calorie content. By leveraging existing technology, specifically the YOLO model, the research aims to detect countable solid foods and calculate their caloric values using a specialized dataset. Different YOLO configurations are evaluated to determine their accuracy in detecting and classifying food items. Federated learning facilitates decentralized training on local datasets, transmitting only model updates to a central server, thus preserving data privacy and meeting regulatory requirements. This method fosters collaborative learning across distributed devices.