Sound Image Generation in 3D Virtual Space Considering Relationship Between Bounding Box by Object Detection in 2D Image and Sound Pressure Level

A system for automatic generation of sound image from a moving or still image has been proposed in the previous studies, and a design principle and issues were discussed. The system detects objects in a moving or still image, a direction and distance for each object are estimated, and it generates a sound image based on the direction and distance. In this paper, a method for sound pressure level controlling is proposed to generate a sound image of an object based on coordinates, width and height of the bounding box obtained by object detection. The proposed method is discussed and demonstrated.