A Nutrient Content Estimation System of Buffet Menu using RGB-D Sensor

To estimate ingested nutritional quantity from a buffet meal in one dish, we have developed an estimation system by using RGB-D sensor. This system uses a RGB-D sensor to get both color image and depth image. First, we extract dish region from captured image. Then, we get height information of food region from dish surface to estimate food volume. By combining meal region separation and meal menu recognition of each region, we can estimate weight of each meal region. By using estimated meal type and its weight, we can estimate the ingested nutritional quantity. As a preliminary experimental result, we can estimate weight of each meal elements within 10% error rate.