Image processing apparatus, moving image encoding apparatus, information processing method and information processing program

• US 20070009157A1
• Filed: 05/31/2006
• Published: 01/11/2007
• Est. Priority Date: 05/31/2005
• Status: Active Grant

• Current Assignee: Fujifilm Corporation (Fujifilm Holdings Corporation)
• Original Assignee: Fuji Photo Film Co Limited (Fujifilm Holdings Corporation)
• Inventors: Shirasaka, Hajime
• Granted Patent: US 8,165,344 B2
• Time in Patent Office: Days
• US Class Current: 382/190
• CPC Class Codes: H04N 19/124 Quantisation
• CPC Class Codes: H04N 19/17 the unit being an image region, e.g. an object
• CPC Class Codes: H04N 19/61 in combination with predictive coding

Abstract

The information processing apparatus comprises: an input unit which inputs moving image data of frames; a reference information acquisition unit which acquires reference information in accordance with the moving image data of one of the frames inputted to the input unit, the reference information forming a reference for identifying a constituent part of a person; a constituent part region extraction unit which extracts, in accordance with the reference information, a constituent part region where the constituent part of the person is situated, from the moving image data of each of the frames inputted after the frame from which the reference information has been acquired by the reference information acquisition unit; a priority region setting unit which sets, in accordance with the constituent part region, a priority region in the moving image data of each of the frames inputted after the frame from which the reference information has been acquired by the reference information acquisition unit; and a processing unit which carries out prescribed information processing preferentially in the priority region set by the priority region setting unit.

Claims

1. An information processing apparatus, comprising:

an input unit which inputs moving image data of frames;

a reference information acquisition unit which acquires reference information in accordance with the moving image data of one of the frames inputted to the input unit, the reference information forming a reference for identifying a constituent part of a person;

a constituent part region extraction unit which extracts, in accordance with the reference information, a constituent part region where the constituent part of the person is situated, from the moving image data of each of the frames inputted after the frame from which the reference information has been acquired by the reference information acquisition unit;

a priority region setting unit which sets, in accordance with the constituent part region, a priority region in the moving image data of each of the frames inputted after the frame from which the reference information has been acquired by the reference information acquisition unit; and

a processing unit which carries out prescribed information processing preferentially in the priority region set by the priority region setting unit.

2. The information processing apparatus as defined in claim 1, further comprising:

a face determination unit which determines a face region of the person from the moving image data of one of the frames inputted to the input unit,

wherein the reference information acquisition unit acquires the reference information from the face region of the person determined by the face determination unit.

3. The information processing apparatus as defined in claim 2, wherein the reference information acquisition unit acquires color information of the face region of the person as the reference information.

4. The information processing apparatus as defined in claim 3, wherein:

the face determination unit determines at least one of regions of eyes, mouth and nose of the person from the moving image data of one of the frames inputted to the input unit; and

the reference information acquisition unit acquires the color information from a part of the face region excluding the at least one of the regions of eyes, mouth and nose of the person, as the reference information.

5. The information processing apparatus as defined in claim 3, wherein the reference information includes a color range specified in accordance with the color information acquired by the reference information acquisition unit.

6. The information processing apparatus as defined in claim 5, wherein the constituent part region extraction unit extracts a pixel region having color information falling within the specified color range, as the constituent part region.

7. The information processing apparatus as defined in claim 1, wherein the priority region setting unit sets a region containing the constituent part region as the priority region.

8. The information processing apparatus as defined in claim 1, further comprising:

a position determination unit which determines at least one of positions of eyes, mouth and nose of the person from the moving image data of one of the frames inputted to the input unit,

wherein the priority region setting unit sets the priority region in accordance with the at least one of the positions of eyes, mouth and nose of the person.

9. A moving image encoding apparatus, comprising:

an input unit which inputs moving image data of frames;

a compression unit which compresses the moving image data of the frames inputted to the input unit;

a quantization unit which quantizes the moving image data of the frames compressed by the compression unit;

an output unit which outputs the moving image data of the frames quantized by the quantization unit;

a reference information acquisition unit which acquires reference information in accordance with the moving image data of one of the frames inputted to the input unit, the reference information forming a reference for identifying a constituent part of a person;

a constituent part region extraction unit which extracts, in accordance with the reference information, a constituent part region where the constituent part of the person is situated, from the moving image data of each of the frames inputted after the frame from which the reference information has been acquired by the reference information acquisition unit; and

a high-level quantization region setting unit which sets, in accordance with the constituent part region, a high-level quantization region in the moving image data of each of the frames inputted after the frame from which the reference information has been acquired by the reference information acquisition unit,

wherein the quantization unit adjusts an adaptive level of quantization to become finer in a block of the high-level quantization region.

10. An information processing method, comprising:

an input step of inputting moving image data of frames;

a reference information acquisition step of acquiring reference information in accordance with the moving image data of one of the frames inputted in the input step, the reference information forming a reference for identifying a constituent part of a person;

a constituent part region extraction step of extracting, in accordance with the reference information, a constituent part region where the constituent part of the person is situated, from the moving image data of each of the frames inputted after the frame from which the reference information has been acquired in the reference information acquisition step;

a priority region setting step of setting, in accordance with the constituent part region, a priority region in the moving image data of each of the frames inputted after the frame from which the reference information has been acquired in the reference information acquisition step; and

a processing step of carrying out prescribed information processing preferentially in the priority region set in the priority region setting step.

11. A computer readable medium having embodied thereon a program for performing information processing by a computer, the program comprising:

a first code segment for an input step of inputting moving image data of frames;

a second code segment for a reference information acquisition step of acquiring reference information in accordance with the moving image data of one of the frames inputted in the input step, the reference information forming a reference for identifying a constituent part of a person;

a third code segment for a constituent part region extraction step of extracting, in accordance with the reference information, a constituent part region where the constituent part of the person is situated, from the moving image data of each of the frames inputted after the frame from which the reference information has been acquired in the reference information acquisition step;

a fourth code segment for a priority region setting step of setting, in accordance with the constituent part region, a priority region in the moving image data of each of the frames inputted after the frame from which the reference information has been acquired in the reference information acquisition step; and

a fifth code segment for a processing step of carrying out prescribed information processing preferentially in the priority region set in the priority region setting step.