Concept for determining a measure for a distortion change in a synthesized view due to depth map modifications

1. An encoder for coding a video comprising:

a depth encoding mechanism configured to encode, using a processor, a depth map associated with a view of a video; and

a distortion measurement mechanism configured to determine, using a processor, a distortion change of a first view of a video synthesized from a second view of the video, the distortion change being caused by a modification to a depth map of the second view, wherein the distortion measurement mechanism is configured to:

obtain first and second synthesis states of the first view, the first synthesis state corresponding to a synthesis of the first view based on the depth map of the second view comprising a first portion in a coded state and a second portion in a non-coded state, and the second synthesis state corresponding to a synthesis of the first view based on the depth map of the second view comprising the first portion and a currently encoded portion of the second portion both in the coded state, and

determine the distortion change based on first and second distortion measures with respect to the first and second synthesis states of the first view, respectively.

2. The encoder according to claim 1 , wherein the first and second distortion measures are determined relative to an undistorted version of the first view, and the distortion measurement mechanism is configured to obtain the undistorted version of the first view in accordance with a synthesis of the first view from the second view based on the depth map of the second view in the non-coded state.

3. The encoder according to claim 1 , wherein the distortion measurement mechanism is configured to obtain and determine with regard to a section of the first view within which changes occur between the first synthesis state and the second synthesis state.

4. The encoder according to claim 1 , wherein the distortion measurement mechanism is configured to use a per-pixel difference measure for determining the distortion of the respective synthesis state.

5. The encoder according to claim 1 , wherein the distortion measurement mechanism is configured to warp texture samples of the currently coded portion from the second view into the first view using the coded state of the depth map of the second view and determine the distortion of the second synthesis state of the first view based the warped texture samples.

6. The encoder according to claim 5 , wherein the distortion measurement mechanism is configured to, interpolate the warped texture samples onto sample positions of the first view.

7. A decoder for decoding a video comprising:

a depth decoding mechanism configured to decode, using a processor, a depth map associated with a first view of a video,

wherein information related to the first view is used to determine a distortion change of a second view of the video, the second view being synthesized from the first view of the video, the distortion change being caused by a modification to the depth map of the first view, wherein the distortion change of the second view is determined at least by:

obtaining first and second synthesis states of the second view, the first synthesis state corresponding to a synthesis of the second view based on the depth map of the first view comprising a first portion in a coded state and a second portion in a non-coded state, and the second synthesis state corresponding to a synthesis of the second view based on the depth map of the first view comprising the first portion and a currently encoded portion of the second portion both in the coded state, and

determining the distortion change based on first and second distortion measures with respect to the first and second synthesis states of the second view, respectively.

8. The decoder according to claim 7 , wherein the first and second distortion measures are determined relative to an undistorted version of the second view, and the undistorted version of the second view is obtained in accordance with a synthesis of the second view based on the depth map of the first view in the non-coded state.

9. The decoder according to claim 7 , wherein the steps of obtaining and determining are performed with regard to a section of the second view within which changes occur between the first synthesis state and the second synthesis state.

10. The decoder according to claim 7 , wherein the step of determining includes use a per-pixel difference measure for determining the distortion of the respective synthesis state.

11. The decoder according to claim 7 , wherein the distortion change of the second view is determined further by warping texture samples of the currently encoded portion from the first view into the second view using the coded state of the depth map of the first view and determining the distortion of the second synthesis state of the second view based on the warped texture samples.

12. The decoder according to claim 11 , wherein the warped texture samples are interpolated onto sample positions of the second view.

13. The decoder according to claim 12 , wherein the interpolation includes hole filling or blending with another view of the video.

14. A method of decoding a video comprising:

decoding, using a processor, a depth map associated with a first view of a video; and

determining, based on information related to the first view, a distortion change of a second view of the video, the second view being synthesized from the first view of the video, the distortion change being caused by a modification to the depth map of the first view, wherein the determining the distortion change of the second view comprises:

obtaining first and second synthesis states of the second view, the first synthesis state corresponding to a synthesis of the second view based on the depth map of the first view comprising a first portion in a coded state and a second portion in a non-coded state, and the second synthesis state corresponding to a synthesis of the second view based on the depth map of the first view comprising the first portion and a currently encoded portion of the second portion in the coded state, and

determining the distortion change based on first and second distortion measures with respect to the first and second synthesis states of the second view, respectively.

15. The method according to claim 14 , wherein the first and second distortion measures are determined relative to an undistorted version of the second view, and the method further comprising obtaining the undistorted version of the second view in accordance with a synthesis of the second view based on the depth map of the first view in the non-coded state.

16. The method according to claim 14 , wherein the steps of obtaining and determining are performed with regard to a section of the second view within which changes occur between the first synthesis state and the second synthesis state.

17. The method according to claim 14 , wherein the step of determining includes use a per-pixel difference measure for determining the distortion of the respective synthesis state.

18. The method according to claim 14 , wherein the determining the distortion change further comprises warping texture samples of the currently encoded portion from the first view into the second view using the coded state of the depth map of the first view, and determining the distortion of the second synthesis state of the second view based on the warped texture samples.

19. The method according to claim 18 , further comprising interpolating the warped texture samples onto sample positions of the second view.

20. A non-transitory computer-readable medium for storing instructions, which when executed by a processor, generates data associated with a video, comprising:

a data stream comprising a depth map associated with a first view of a video, the instructions comprising:

encoding the depth map associated with the first view; and

determining a distortion change of a second view of the video, the second view being synthesized from the first view of the video and the distortion change being caused by a modification to the depth map of the first view, and the determining including:

obtaining first and second synthesis states of the second view, the first synthesis state corresponding to a synthesis of the second view based on the depth map of the first view comprising a first portion in a coded state and a second portion in a non-coded state, and the second synthesis state corresponding to a synthesis of the second view based on the depth map of the first view comprising the first portion and a currently encoded portion of the second portion in the coded state, and

determining the distortion change based on first and second distortion measures with respect to the first and second synthesis states of the second view, respectively.