Can AI determine the perfect clothing sizes from a series of photos ?

Current AI systems can estimate basic body measurements from single photos with moderate accuracy, but they do not yet determine “perfect” clothing sizes that account for fabric drape, brand-specific fit rules, or individual comfort preferences. Most commercial tools rely on 2D pose estimation and anthropometric models to infer height, bust, waist, and hip dimensions, achieving typical errors of ±2–3 cm in controlled settings. More advanced systems combine multiple views or short videos to reduce occlusion and improve volumetric reconstruction, yet they still output static measurements rather than a curated size recommendation. Fully automated “perfect fit” determination remains out of reach because it requires real-time integration of material properties, user feedback, and retailer-specific grading standards. SOURCE: McKinsey & Company — https://www.mckinsey.com/industries/retail/our-insights/notes-from-the-ai-frontier-modeling-the-impact-of-ai-on-the-world-economy

— Enriched May 13, 2026

AI can estimate clothing sizes from photos with varying degrees of accuracy, using computer vision and machine learning techniques to analyze body measurements and proportions. However, determining the perfect clothing size remains a challenging task due to factors such as pose, lighting, and clothing type. Current systems often require multiple photos from different angles and may still produce inaccurate results, highlighting the need for further research and development. While progress has been made, achieving high accuracy and reliability in clothing size estimation from photos is an ongoing effort.

— Enriched May 13, 2026 · Source: IEEE