3D Food
3D food modeling is an emerging field focused on creating realistic three-dimensional representations of food items for applications in nutrition monitoring, dietary assessment, and robotics. Current research emphasizes developing accurate and efficient methods for 3D food reconstruction from 2D images, often employing techniques like generative models and structure-from-motion, and incorporating nutritional information directly into the models. This work is driven by the need for improved data sets and algorithms to enable more precise portion estimation and ultimately enhance personalized dietary management and robotic food assistance. The resulting advancements have significant implications for improving public health and assistive technologies.
Papers
NutritionVerse-Thin: An Optimized Strategy for Enabling Improved Rendering of 3D Thin Food Models
Chi-en Amy Tai, Jason Li, Sriram Kumar, Saeejith Nair, Yuhao Chen, Pengcheng Xi, Alexander Wong
NutritionVerse-3D: A 3D Food Model Dataset for Nutritional Intake Estimation
Chi-en Amy Tai, Matthew Keller, Mattie Kerrigan, Yuhao Chen, Saeejith Nair, Pengcheng Xi, Alexander Wong