Object condition judgment system, object condition judgment method, and program

摘要

PROBLEM TO BE SOLVED: To perform efficient learning without preparing a large number of images and to infer the situation of an object in a new image based on the learning result. And provide programs. SOLUTION: A component related to at least two of an object captured in an image, a posture, shape, direction, expression of the object, a type and position of a part of the object, and a background of the captured image are included in the image. A component analysis module 111 that extracts and analyzes feature points, and a situation in which status data on the situation of an object is acquired from an image search computer that acquires characters and searches for images related to the characters. The acquisition module 112, the learning module 113 that learns by associating the acquired situation data with the combination of the components, and the result analyzed by the component analysis module 111 for a predetermined image are similar to the combination of the components. Alternatively, in the same case, the object situation estimation module 114 for estimating the situation of the object based on the learning result of the situation data is provided. [Selection diagram] Fig. 1