Object identification method, device and program using coded parameter features

摘要

PROBLEM TO BE SOLVED: To provide an object identification method capable of identifying an object accompanied by movement or change with a smaller processing load. The object identification method is an object identification method in a computer that identifies an object from image data that can include a predetermined object in an image, and is a unit image when video data including the image data is encoded. Generates a coding parameter feature amount, which is a feature amount of mapping information obtained by mapping a coding parameter determined for each area to the unit image area, and an image feature amount, which is a feature amount related to a pixel value of image data. And the step of identifying the target from the image data using a trained classifier that inputs the generated coding parameter feature amount and image feature amount and outputs the information related to the predetermined target class. Has. Here, the coding of the video data including the image data may be the coding by MPEG, and the unit image area can be a macroblock. [Selection diagram] Fig. 1