Human detection device and human detection method

A human model fitting device which causes a human movement model defining a movement characteristic of a human to be fitted to a movement of a human in an image, said device comprising:

• a spatiotemporal volume generation unit configured to generate a three-dimensional spatiotemporal image in which frame images that make up the moving picture in which a human has been filmed are arranged along a temporal axis;
• a spatiotemporal fragment extraction unit configured to extract a real image spatiotemporal fragment, which is an image appearing in a cut plane or cut fragment when the three-dimensional spatiotemporal image is cut, from the generated three-dimensional spatiotemporal image;
• a spatiotemporal fragment output unit configured to generate and output a human body region movement spatiotemporal fragment, which is a spatiotemporal fragment obtained from a movement by the human movement model;
• a spatiotemporal fragment verification unit configured to verify a real image spatiotemporal fragment extracted by said spatiotemporal fragment extraction unit with a human body region movement model spatiotemporal fragment outputted by said spatiotemporal fragment output unit;
• a model fitting unit configured to determine a value of parameters which specifies the human movement model so that the human movement model indicates a movement of a human occurring in the moving picture, based on a verification result from said spatiotemporal fragment verification unit;
• a display unit configured to display the presence/absence of a human based on an output of said model fitting unit.

A human detection method for detecting a human contained in a moving picture, wherein said method comprises using a computer processor to perform the steps of:

• generating, using a spatiotemporal volume generation unit, a three-dimensional spatiotemporal image in which frame images that make up the moving picture in which a human has been filmed are arranged along a temporal axis;
• extracting, using a spatiotemporal fragment extraction unit, and from the generated three-dimensional spatiotemporal image, a real image spatiotemporal fragment, which is an image appearing in a cut plane or cut fragment when the three-dimensional spatiotemporal image is cut;
• outputting, using a spatiotemporal fragment output unit, and based on a human movement model which defines a characteristic of the movement of a human, a human body region movement spatiotemporal fragment, which is a spatiotemporal fragment obtained from a movement by the human movement model;
• verifying, using a spatiotemporal fragment verification unit, a real image spatiotemporal fragment extracted in said extracting with a human body region movement model spatiotemporal fragment outputted in said outputting;
• and
• outputting, using an attribute output unit, a human attribute which includes a presence/absence of a human in the moving picture, based on a verification result of said verifying.

A human model fitting method which causes a human movement model defining a movement characteristic of a human to be fitted to a movement of a human in an image, wherein said method comprises using a computer processor to performs the steps of:

• generating, using a spatiotemporal volume generation unit, a three-dimensional spatiotemporal image in which frame images that make up the moving picture in which a human has been filmed are arranged along a temporal axis;
• extracting, using a spatiotemporal fragment extraction unit, a real image spatiotemporal fragment, which is an image appearing in a cut plane or cut fragment when the three-dimensional spatiotemporal image is cut, from the generated three-dimensional spatiotemporal image;
• generating and outputting, using a spatiotemporal fragment output unit, a human body region movement spatiotemporal fragment, which is a spatiotemporal fragment obtained from a movement by the human movement model;
• verifying, using a spatiotemporal fragment verification unit, a real image spatiotemporal fragment extracted in said extracting with a human body region movement model spatiotemporal fragment outputted in said outputting;
• and
• determining, using a model fitting unit, a value of parameters which specifies the human movement model so that the human movement model indicates a movement of a human occurring in the moving picture, based on a verification result of said verification.