Method and system for video transcoding based on spatial or temporal importance

1. A non-transitory computer-readable storage medium storing a set of instructions that are executable by one or more processors of a device to cause the device to perform a method comprising:

in response to receiving an encoded video bitstream, decoding a picture from the encoded video bitstream;

determining a value representing a first level of spatial importance for a first region of a background of the picture based on an image segmentation technique;

determining a value representing a second level of spatial importance for a second region of the picture based on the image segmentation technique, wherein the second level of spatial importance is higher than the first level of spatial importance, and the second region is part of one of the background or a region of interest of the picture;

applying to the first region a first resolution-enhancement technique associated with the first level of spatial importance for increasing resolution of the first region by a scaling factor, wherein the first resolution-enhancement technique is selected from a set of resolution-enhancement techniques having different computational complexity levels;

applying to the second region a second resolution-enhancement technique associated with the second level of spatial importance for increasing resolution of the second region by the scaling factor, wherein a computational complexity of the second resolution-enhancement technique is higher than a computational complexity of the first resolution-enhancement technique; and

encoding the first region and the second region using a video coding standard.

2. The non-transitory computer-readable storage medium of claim 1 , wherein encoding the second region further comprises:

encoding the second region using a quantization parameter having a value smaller than or equal to a value of a quantization parameter used for encoding the first region.

3. The non-transitory computer-readable storage medium of claim 1 , wherein encoding the second region further comprises:

determining whether the second region is decoded using intra prediction or inter prediction; and

based on a determination that the second region is decoded using the inter prediction, encoding the second region using an inter prediction mode of the video coding standard, wherein a search window of motion estimation for a basic processing sub-unit in the second region is a region of interest in a reference picture associated with the second region.

4. The non-transitory computer-readable storage medium of claim 1 , wherein the second resolution-enhancement technique is a learning-based technique, and the first resolution-enhancement technique is one of an interpolation-based technique, a reconstruction-based technique, or a null operation that does not perform any resolution enhancement.

5. The non-transitory computer-readable storage medium of claim 4 , wherein applying to the first region the first resolution-enhancement technique further comprises:

determining whether the first region is decoded using intra prediction or inter prediction; and

based on a determination that the first region is decoded using the intra prediction, applying to the first region one of the interpolation-based technique or the reconstruction-based technique.

6. The non-transitory computer-readable storage medium of claim 5 , wherein the set of instructions that are executable by the one or more processors cause the device to further perform:

based on a maximum computational complexity allowed by the device, selecting one of the reconstruction-based technique or the interpolation-based technique for applying to the first region.

7. The non-transitory computer-readable storage medium of claim 5 , wherein the set of instructions that are executable by the one or more processors cause the device to further perform:

based on a determination that the first region is decoded using the inter prediction, determining whether any basic processing sub-unit of the first region has a size undefined in the video coding standard after the resolution of the first region is increased by the scaling factor;

based on a determination that no basic processing sub-unit of the first region has the size undefined in the video coding standard after the resolution of the first region is increased by the scaling factor, applying to the first region the null operation; and

updating encoding data associated with the first region in accordance with the scaling factor, wherein the encoding data is decoded from the encoded video bitstream.

8. The non-transitory computer-readable storage medium of claim 7 , wherein the set of instructions that are executable by the one or more processors cause the device to further perform:

based on a determination that at least one basic processing sub-unit of the first region has the size undefined in the video coding standard after the resolution of the first region is increased by the scaling factor, applying to the first region one of the interpolation-based technique or the reconstruction-based technique.

9. The non-transitory computer-readable storage medium of claim 7 , wherein encoding the first region further comprises:

encoding the updated encoding data using an inter prediction mode of the video coding standard.

10. The non-transitory computer-readable storage medium of claim 9 , wherein encoding the updated encoding data further comprises:

encoding the updated encoding data using a quantization parameter having a value higher than or equal to a value of a quantization parameter used by the encoded video bitstream to encode the first region.

11. The non-transitory computer-readable storage medium of claim 7 , wherein the encoding data comprises at least one of decoded coordinates of a matching region associated with the first region, a decoded motion vector associated with the matching region, or a decoded residual region associated with the first region.

12. The non-transitory computer-readable storage medium of claim 11 , wherein updating the encoding data further comprises:

upscaling, by the scaling factor, a location and a size of the decoded coordinates of the matching region, and a location and a magnitude of the decoded motion vector;

determining whether any pixel of the decoded residual region has a value exceeding a predetermined threshold value;

based on a determination that no pixel of the decoded residual region has a value exceeding the predetermined threshold value, enlarging the decoded residual region by the scaling factor; and

based on a determination that at least one pixel of the decoded residual region has a value exceeding the predetermined threshold value, determining a residual region based on the upscaled matching region and upscaled motion vector.

13. The non-transitory computer-readable storage medium of claim 7 , wherein basic processing sub-unit comprises one of a block or a coding unit.

14. The non-transitory computer-readable storage medium of claim 1 , wherein the image segmentation technique comprises one of a semantic segmentation technique or an instance segmentation technique.

15. An apparatus, comprising:

a memory configured to store a set of instructions; and

one or more processors coupled to the memory and configured to execute the set of instructions to cause the apparatus to:

in response to receiving an encoded video bitstream, decode a picture from the encoded video bitstream;

determine a value representing a first level of spatial importance for a first region of a background of the picture based on an image segmentation technique;

determine a value representing a second level of spatial importance for a second region of the picture based on the image segmentation technique, wherein the second level of spatial importance is higher than the first level of spatial importance, and the second region is part of one of the background or a region of interest of the picture;

apply to the first region a first resolution-enhancement technique associated with the first level of spatial importance for increasing resolution of the first region by a scaling factor, wherein the first resolution-enhancement technique is selected from a set of resolution-enhancement techniques having different computational complexity levels;

apply to the second region a second resolution-enhancement technique associated with the second level of spatial importance for increasing resolution of the second region by the scaling factor, wherein a computational complexity of the second resolution-enhancement technique is higher than a computational complexity of the first resolution-enhancement technique; and

encode the first region and the second region using a video coding standard.

16. The apparatus of claim 15 , wherein the second resolution-enhancement technique is a learning-based technique, and the first resolution-enhancement technique is one of an interpolation-based technique, a reconstruction-based technique, or a null operation that does not perform any resolution enhancement.

17. The apparatus of claim 15 , wherein the one or more processors configured to execute the set of instructions to cause the apparatus to apply to the first region the first resolution-enhancement technique are further configured to execute the set of instructions to cause the apparatus to:

determine whether the region is decoded using intra prediction or inter prediction; and

based on a determination that the region is decoded using the intra prediction, apply to the first region one of the interpolation-based technique or the reconstruction-based technique.

18. The apparatus of claim 17 , wherein the one or more processors are further configured to execute the set of instructions to cause the apparatus to:

based on a determination that the region is decoded using the inter prediction, determine whether any basic processing sub-unit of the first region has a size undefined in the video coding standard after the resolution of the first region is increased by the scaling factor;

based on a determination that no basic processing sub-unit of the first region has the size undefined in the video coding standard after the resolution of the first region is increased by the scaling factor, apply to the first region the null operation; and

update encoding data associated with the first region in accordance with the scaling factor, wherein the encoding data is decoded from the encoded video bitstream.

19. The apparatus of claim 18 , wherein the one or more processors are further configured to execute the set of instructions to cause the apparatus to:

based on a determination that at least one basic processing sub-unit of the first region has the size undefined in the video coding standard after the resolution of the first region is increased by the scaling factor, apply to the first region one of the interpolation-based technique or the reconstruction-based technique.

20. The apparatus of claim 18 , wherein the one or more processors configured to execute the set of instructions to cause the apparatus to encode the first region are further configured to execute the set of instructions to cause the apparatus to:

encode the updated encoding data using an inter prediction mode of the video coding standard.

21. The apparatus of claim 18 , wherein the encoding data comprises at least one of decoded coordinates of a matching region associated with the first region, a decoded motion vector associated with the matching region, or a decoded residual region associated with the first region.

22. The apparatus of claim 21 , wherein the one or more processors configured to execute the set of instructions to cause the apparatus to update the encoding data are further configured to execute the set of instructions to cause the apparatus to:

upscale, by the scaling factor, a location and a size of the decoded coordinates of the matching region, and a location and a magnitude of the decoded motion vector;

determine whether any pixel of the decoded residual region has a value exceeding a predetermined threshold value;

based on a determination that no pixel of the decoded residual region has a value exceeding the predetermined threshold value, enlarge the decoded residual region by the scaling factor; and

based on a determination that at least one pixel of the decoded residual region has a value exceeding the predetermined threshold value, determine a residual region based on the upscaled matching region and upscaled motion vector.

23. A computer-implemented method, comprising:

in response to receiving an encoded video bitstream, decoding a picture from the encoded video bitstream;

determining a value representing a first level of spatial importance for a first region of a background of the picture based on an image segmentation technique;

determining a value representing a second level of spatial importance for a second region of the picture based on the image segmentation technique, wherein the second level of spatial importance is higher than the first level of spatial importance, and the second region is part of one of the background or a region of interest of the picture;

applying to the first region a first resolution-enhancement technique associated with the first level of spatial importance for increasing resolution of the first region by a scaling factor, wherein the first resolution-enhancement technique is selected from a set of resolution-enhancement techniques having different computational complexity levels;

applying to the second region a second resolution-enhancement technique associated with the second level of spatial importance for increasing resolution of the second region by the scaling factor, wherein a computational complexity of the second resolution-enhancement technique is higher than a computational complexity of the first resolution-enhancement technique; and

encoding the first region and the second region using a video coding standard.

24. The computer-implemented method of claim 23 , wherein the second resolution-enhancement technique is a learning-based technique, and the first resolution-enhancement technique is one of an interpolation-based technique, a reconstruction-based technique, or a null operation that does not perform any resolution enhancement.

25. The computer-implemented method of claim 23 , wherein applying to the first region the first resolution-enhancement technique comprises:

determining whether the region is decoded using intra prediction or inter prediction; and

based on a determination that the region is decoded using the intra prediction, applying to the first region one of the interpolation-based technique or the reconstruction-based technique.

26. The computer-implemented method of claim 25 , further comprising:

based on a determination that the region is decoded using the inter prediction, determining whether any basic processing sub-unit of the first region has a size undefined in the video coding standard after the resolution of the first region is increased by the scaling factor;

based on a determination that no basic processing sub-unit of the first region has the size undefined in the video coding standard after the resolution of the first region is increased by the scaling factor, applying to the first region the null operation; and

updating encoding data associated with the first region in accordance with the scaling factor, wherein the encoding data is decoded from the encoded video bitstream.

27. The computer-implemented method of claim 26 , further comprising:

based on a determination that at least one basic processing sub-unit of the first region has the size undefined in the video coding standard after the resolution of the first region is increased by the scaling factor, applying to the first region one of the interpolation-based technique or the reconstruction-based technique.

28. The computer-implemented method of claim 26 , wherein encoding the first region comprises:

encoding the updated encoding data using an inter prediction mode of the video coding standard.

29. The computer-implemented method of claim 26 , wherein the encoding data comprises at least one of decoded coordinates of a matching region associated with the first region, a decoded motion vector associated with the matching region, or a decoded residual region associated with the first region.

30. The computer-implemented method of claim 29 , wherein updating the encoding data comprises:

upscaling, by the scaling factor, a location and a size of the decoded coordinates of the matching region, and a location and a magnitude of the decoded motion vector;

determining whether any pixel of the decoded residual region has a value exceeding a predetermined threshold value; and

based on a determination that no pixel of the decoded residual region has a value exceeding the predetermined threshold value, enlarging the decoded residual region by the scaling factor; or

based on a determination that at least one pixel of the decoded residual region has a value exceeding the predetermined threshold value, determining a residual region based on the upscaled matching region and upscaled motion vector.