Method and apparatus to encode image, and method and apparatus to decode image data

1. A method of encoding an image, comprising:

generating a first bitstream by encoding an omni-directional image and transmitting the first bitstream to a decoding apparatus;

receiving from the decoding apparatus position information of a region-of-interest selected from an image reconstructed based on the first bitstream; and

generating a second bitstream by encoding an image of the region-of-interest based on the position information,

wherein the generating of the second bitstream comprises:

obtaining a first image corresponding to the region-of-interest from the omni-directional image;

obtaining a reconstructed omni-directional image by decoding the first bitstream;

obtaining a second image corresponding to the region-of-interest from the reconstructed omni-directional image; and

obtaining an error image corresponding to a difference between the first image and the second image and encoding the error image to generate the second bitstream.

2. The method of claim 1 , wherein

the obtaining of the first image comprises:

converting the omni-directional image into a first panorama image, and

obtaining the first image from the first panorama image, and

the obtaining of the second image comprises:

converting the reconstructed omni-directional image into a second panorama image, and

obtaining the second image from the second panorama image.

3. The method of claim 2 , wherein

the obtaining of the first image from the first panorama image comprises:

selecting the region-of-interest from the first panorama image, and

converting the region-of-interest selected from the first panorama image into perspective image to obtain the first image, and

the obtaining of the second image from the second panorama image comprises:

selecting the region-of-interest from the second panorama image, and

converting the region-of-interest selected from the second panorama image into a perspective image to obtain the second image.

4. The method of claim 1 , further comprising:

down-sampling the omni-directional image before generating the first bitstream, wherein the obtaining of the reconstructed omni-directional image comprises performing up-sampling corresponding to the down-sampling.

5. A method of encoding an image, comprising:

encoding an input omni-directional image and outputting the encoded omni-directional image at a first resolution;

encoding an image corresponding to a region of the input omni-directional image determined according to input position information and outputting the encoded image corresponding to the region of the input omni-directional image at a second resolution,

wherein the encoding of the image corresponding to the region of the input omni-directional image comprises:

decoding the encoded omni-directional image output at the first resolution;

generating a first perspective image of a region of the decoded omni-directional image according to the input position information;

generating a second perspective image of the region of the input omni-directional image according to the input position information;

calculating an error image corresponding to a difference between the second perspective image and the first perspective image; and

encoding the calculated error image.

6. The method of claim 5 , wherein the second resolution is higher than the first resolution.

7. An apparatus to encode an image, comprising:

a first encoder to encode an omni-directional image to generate a first bitstream;

a data communicator to transmit the first bitstream to a decoding apparatus and to receive from the decoding apparatus position information of a region-of-interest selected from an image reconstructed based on the first bitstream;

a region-of-interest selector to receive the position information of the region-of-interest and to output a region selection control signal;

a first conversion unit to output a first image corresponding to the region-of-interest in the omni-directional image in response to the region selection control signal;

a second conversion unit to output a second image corresponding to the region-of-interest in a reconstructed omni-directional image, which is generated by the first encoder by decoding the first bitstream, in response to the region selection control signal; and

a subtractor to output an error image between the first image and the second image to the second encoder as the image of the region-of-interest to be encoded by the second encoder; and

a second encoder to encode an image of the region-of-interest based on the position information to generate a second bitstream.

8. The apparatus of claim 7 , wherein

the first conversion unit comprises:

a first panorama image generator to convert the omni-directional image into a first panorama image and to output the first panorama Image, and

a first perspective image generator to convert a portion corresponding to the region-of-interest in the first panorama image into a first perspective image and to output the first perspective image, and

the second conversion unit comprises:

a second panorama image generator to convert the reconstructed omni-directional image into a second panorama image and to output the second panorama image, and

a second perspective image generator to convert a portion corresponding to the region-of-interest in the second panorama image into a second perspective image and to output the second perspective image.

9. The apparatus of claim 7 , further comprising:

a down-sampler to down-sample the omni-directional image and to output the result of the down-sampling to the first encoder; and

an up-sampler to perform up-sampling corresponding to the down-sampling with respect to the reconstructed omni-directional image generated by the first encoder and to output the result of the up-sampling.

10. An image encoding apparatus, comprising:

a first encoding unit to encode an input omni-directional image and to decode the encoded input omni-directional image;

a region-of-interest unit to receive input position information of a selected region of interest of the input omni-directional image and to generate first and second perspective images of the selected region of interest in the input omni-directional image and the decoded omni-directional image, respectively, according to the received position information, the region-of-interest unit including:

a first conversion unit to generate the first perspective image of the selected region in the input omni-directional image;

a second conversion unit to generate the perspective second image of the selected region in the decoded omni-directional image: and

a region-of-interest selector to receive the input position information of the selected region of interest and to control the first and second conversion units according to the received position information; and

a second encoding unit to encode an error image corresponding to a difference between the first and second perspective images of the selected region of interest,

wherein the second encoding unit transmits the encoded error image at a higher resolution than the first encoding unit transmits the encoded omni-directional image.

11. The image encoding apparatus of claim 10 , wherein the first conversion unit comprises a first omni-to-panorama converter to convert the input omni-directional image into a first panorama image and a first panorama-to-perspective image converter to convert a portion of the first panorama image into the first perspective image according to the input position information, and the second conversion unit comprises a second omni-to-panorama converter to convert the decoded omni-directional image into a second panorama image and a second panorama-to-perspective image converter to convert a portion of the second panorama image into the second perspective image according to the input position information.

12. The image encoding apparatus of claim 10 , further comprising:

a data communicator to transmit the encoded omni-directional image and the encoded error image to an external decoding apparatus and to receive the input position information of the selected region-of-interest from the external decoding apparatus and transmit the received position information to the region-of-interest unit.

13. The image encoding apparatus of claim 10 , wherein the region-of-interest is selected from the decoded omni-directional image.

14. A method of decoding an image, comprising:

receiving from an encoding apparatus a first bitstream generated by encoding an omni-directional image;

decoding the first bitstream and displaying a reconstructed image;

transmitting position information of a region-of-interest selected from the displayed reconstructed image to the encoding apparatus;

receiving a second bitstream generated by encoding an image of the region-of-interest from the encoding apparatus; and

decoding the second bitstream,

wherein the decoding of the second bitstream comprises:

obtaining a first image corresponding to the region-of-interest by decoding the second bitstream;

obtaining a second image corresponding to the region-of-interest from the reconstructed image; and

generating a combined perspective image corresponding to the region-of-interest by mixing the first image and the second image.

15. The method of claim 14 , wherein the obtaining of the second image comprises:

converting the reconstructed image into a panorama image;

selecting the region-of-interest from the panorama image; and

converting the region-of-interest into a perspective image to obtain the second image.

16. The method of claim 14 , further comprising, when down-sampling is performed when the encoding apparatus encodes the omni-directional image, performing up-sampling corresponding to the down-sampling with respect to the reconstructed omni-directional image.

17. An apparatus to decode an image, comprising:

a first decoder to receive from an encoding apparatus a first bitstream generated by encoding an omni-directional image and to decode the first bitstream to generate a reconstructed omni-directional image;

a first display unit to display the reconstructed omni-directional image decoded by the first decoder;

a data communicator to transmit position information of a region-of-interest selected from the reconstructed omni-directional image displayed by the first display unit to the encoding apparatus;

a second decoder to receive a second bitstream generated by encoding an image of the region-of-interest from the encoding apparatus and to decode the second bitstream;

a region-of-interest selector to receive the position information of the selected region-of-interest and to output a region selection control signal;

a conversion unit to output an image corresponding to the region-of-interest in the reconstructed omni-directional image In response to the region selection control signal; and

a mixer to mix an image output from the second decoder and the image output from the conversion unit.

18. The apparatus of claim 17 , wherein the conversion unit comprises:

a panorama image generator to convert the reconstructed omni-directional image decoded by the first decoder into a panorama image; and

a perspective image generator to convert a portion corresponding to the region-of-interest in the panorama image into a perspective image in response to the region selection control signal and to output the perspective image to the mixer.

19. The apparatus of claim 17 , further comprising:

an up-sampler to perform up-sampling corresponding to down-sampling performed when the encoding apparatus encodes the omni-directional image with respect to the reconstructed omni-directional image.

20. An image decoding apparatus, comprising:

a first decoding unit to decode an input encoded omni-directional image;

a region-of-interest unit to select a region-of-interest in the decoded omni-directional image, to output position information of the selected region of interest, and to generate a perspective image of the selected region-of-interest in the decoded omni-directional image;

a second decoding unit to decode an input error image; and

a calculating unit to combine the perspective image of the selected region-of-interest in the decoded omni-directional image and the decoded error image to form a combined region-of-interest image,

wherein the region-of-interest unit comprises:

a display to display the decoded omni-directional image;

a user interface to allow a user to select the region-of-interest according to the displayed omni-directional image; and

a conversion unit to generate the perspective image of the selected region-of-interest in the decoded omni-directional image.

21. The image decoding apparatus of claim 20 , wherein the conversion unit comprises:

an omni-to-panorama converter to convert the decoded omni-directional image into a panorama image and output the panorama image to the display to be displayed thereon; and

a panorama-to-perspective image converter to convert a portion of the panorama image corresponding to the selected region-of-interest to the perspective image of the selected region-of-interest.

22. The image decoding apparatus of claim 20 , further comprising:

one or more displays to display the decoded omni-directional image and the combined region-of-interest image.

23. A non-transitory computer readable recording medium to record a program to implement a method of decoding an image, the method comprising:

receiving from an encoding apparatus a first bitstream generated by encoding an omni-directional image;

decoding the first bitstream and displaying a reconstructed image;

transmitting position information of a region-of-interest selected from the reconstructed image to the encoding apparatus;

receiving from the encoding apparatus a second bitstream generated by encoding an image of the region-of-interest; and

decoding the second bitstream,

wherein the decoding of the second bitstream comprises:

obtaining a first image corresponding to the region-of-interest by decoding the second bitstream;

obtaining a second image corresponding to the region-of-interest from the reconstructed image; and

generating a combined perspective image corresponding to the region-of-interest by mixing the first image and the second image.

24. A non-transitory computer readable recording medium to record a program to implement a method of encoding an image, the method comprising:

encoding an input omni-directional image and outputting the encoded omni-directional image at a first resolution; and

encoding an image corresponding to a region of the input omni-directional image determined according to input position information and outputting the encoded image corresponding to the region of the input omni-directional image at a second resolution,

wherein the encoding of the image corresponding to the region of the input omni-directional image comprises:

decoding the encoded omni-directional image output at the first resolution;

generating a first perspective image of a region of the decoded omni-directional image according to the input position information;

generating a second perspective image of the region of the input omni-directional image according to the input position information;

calculating an error image corresponding to a difference between the second perspective image and the first perspective image; and

encoding the calculated error image.