Food Volume Estimation by Integrating 3D Image Projection and Manual Wire Mesh Transformations

2D images can be used to capture food intake data in nutrition studies. Estimates of volume from these are required for nutrient analysis. Although 3D image is possible, it not commonplace. Additionally, studies often require multiple taken by non-expert users, typically collected using mobile phones, due their convenience. Current approaches restricted the need prescribed camera placement, metadata analysis and/or significant computational resources. A new method presented combining and automated scene projection with manual placement resizing wire mesh objects. images, a reference object, on low specification phones. calculated twinning cuboid space object image. manually selected then positioned over target item transformed improve accuracy. The virtual projected into calculated. whole process computationally light, runs real-time as an app standard Apple iPad. Based user study 60 participants, experimental evaluations estimates regular shape ground truth items demonstrate that this approach provides acceptable We accuracy increased independent estimates.