New system estimates calories in Bangladeshi street food using AI models

Researchers have developed a vision-based system for estimating the calorie content of Bangladeshi street food, addressing a gap in current Western-centric approaches. The study compared five object detection and segmentation models, with YOLO11n achieving the highest detection performance. For calorie prediction, a Random Forest regression model, utilizing geometric features extracted by YOLO11n and a Bangladeshi 5 Taka coin for scale reference, yielded the best results with a mean absolute error of 5.68. AI

IMPACT This research could enable more accurate dietary monitoring and the development of specialized mobile health applications for diverse global cuisines.

RANK_REASON The cluster contains an academic paper detailing a new methodology and comparative study of AI models for a specific application. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• AdaBoost
• gradient boosting
• Md Tamim Hossain
• Random Forest
• RF-DETR
• YOLO11n
• YOLO12n
• YOLO26n
• YOLOv8n

• paper
• product
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →