Process and system for encoding and playback of stereoscopic video sequences
View Patent ↗A method for decoding a compressed image stream, the image stream having a plurality of frames, each frame consisting of a merged image including pixels from a left image and pixels from a right image. The method involves the steps of receiving each merged image; changing a clock domain from the original input signal to an internal domain; for each merged image, placing at least two adjacent pixels into an input buffer and interpolating an intermediate pixel, for forming a reconstructed left frame and a reconstructed right frame according to provenance of the adjacent pixels; and reconstructing a stereoscopic image stream from the left and right image frames. The invention also teaches a system for decoding a compressed image stream.
1. A decoder for decoding an encoded stereoscopic digital image stream containing a merged image frame having a first and a second spatially compressed subimage areas corresponding to respective original left and right image frames of an original stereoscopic image stream corresponding to a left-eye view and a right-eye view of a scene respectively,
the first and second spatially compressed subimage areas comprising pixels subsampled from said respective original left and right image frames according to different and complementary respective first and second subsampling coordinates,
such that the pixels of the original left image at the second subsampling coordinates are absent from the first spatially compressed subimage area and the pixels of the original right image at the first subsampling coordinates are absent from the second spatially compressed subimage area,
the decoder comprising at least one processor configured to perform:
reading pixels from said first and second spatially compressed subimage areas of pixels of said digital image stream;
reproducing a decoded left image frame by applying a pixel interpolation to the first spatially compressed subimage area in order to recover decoded left image pixels corresponding to the pixels of the original left image at the second subsampling coordinates;
reproducing a decoded right image frame by applying a pixel interpolation to the second compressed subimage area in order to recover decoded right image pixels corresponding to the pixels of the original right image at the first subsampling coordinates;
outputting a decoded stereoscopic image stream comprising said decoded left image frame and said decoded right image frame respectively corresponding substantially to said left-eye view and said right-eye view of said scene.
2. The decoder as claimed in claim 1 , wherein:
the reproducing a decoded left image frame further comprises placing the pixels of the first spatially compressed subimage area into the decoded left image frame at the first subsampling coordinates and placing the recovered decoded left image pixels in the decoded left image at the second subsampling coordinates; and
the reproducing a decoded right image frame further comprises placing the pixels of the second spatially compressed subimage area into the decoded right image frame at the second subsampling coordinates and placing the recovered decoded right image pixels in the decoded right image at the first subsampling coordinates.
3. The decoder as claimed in claim 1 , wherein said spatially uncompressed image frames have in one dimension a same number of pixels as said spatially compressed subimage areas, and said spatially uncompressed image frames have in an orthogonal dimension twice as many pixels as said spatially compressed subimage areas.
4. The decoder as claimed in claim 3 , wherein said subimages are concatenated side-by-side.
5. The decoder as claimed in claim 2 , wherein said subsampling coordinates follow a checkerboard pattern.
6. The decoder as claimed in claim 3 , wherein said subsampling coordinates follow a checkerboard pattern.
7. The decoder as claimed in claim 1 , wherein said decoding further comprises decoding said digital image stream to obtain said merged video frames.
8. The decoder as claimed in claim 7 , wherein said decoding comprises an MPEG standard decoding of said merged video frames.
9. The decoder as claimed in claim 8 , wherein said decoding comprises MPEG-2 decoding of said merged video frames.
10. A method for decoding an encoded stereoscopic digital image stream containing a merged image frame having a first and a second spatially compressed subimage areas corresponding to respective original left and right image frames of an original stereoscopic image stream corresponding to a left-eye view and a right-eye view of a scene respectively,
the first and second spatially compressed subimage areas comprising pixels subsampled from said respective original left and right image frames according to different and complementary respective first and second subsampling coordinates,
such that the pixels of the original left image at the second subsampling coordinates are absent from the first spatially compressed subimage area and the pixels of the original right image at the first subsampling coordinates are absent from the second spatially compressed subimage area,
the method comprising:
reading pixels from said subimage areas of pixels of said digital image stream; and
reproducing a decoded left image frame by applying a pixel interpolation to the first spatially compressed subimage area in order to recover decoded left image pixels corresponding to the pixels of the original left image at the second subsampling coordinates;
reproducing a decoded right image frame by applying a pixel interpolation to the second compressed subimage area in order to recover decoded right image pixels corresponding to the pixels of the original right image at the first subsampling coordinates;
outputting a decoded stereoscopic image stream comprising said decoded left image frame and said decoded right image frame respectively corresponding substantially to said left-eye view and said right-eye view of said scene.
11. The method as claimed in claim 10 , wherein:
the reproducing a decoded left image frame further comprises placing the pixels of the first spatially compressed subimage area into the decoded left image frame at the first subsampling coordinates and placing the recovered decoded left image pixels in the decoded left image at the second subsampling coordinates; and
the reproducing a decoded right image frame further comprises placing the pixels of the second spatially compressed subimage area into the decoded right image frame at the second subsampling coordinates and placing the recovered decoded right image pixels in the decoded right image at the first subsampling coordinates.
12. The method as claimed in claim 10 , wherein said spatially uncompressed image frames have in one dimension a same number of pixels as said spatially compressed subimage areas, and said spatially uncompressed image frames have in an orthogonal dimension twice as many pixels as said spatially compressed subimage areas.
13. The method as claimed in claim 12 , wherein said subimages are concatenated side-by-side.
14. The method as claimed in claim 11 , wherein said subsampling coordinates follow a checkerboard pattern.
15. The method as claimed in claim 12 , wherein said subsampling coordinates follow a checkerboard pattern.
16. The method as claimed in claim 10 , wherein said decoding further comprises decoding said digital image stream to obtain said merged video frames.
17. The decoder as claimed in claim 16 , wherein said decoding comprises an MPEG standard decoding of said merged video frames.
18. The decoder as claimed in claim 17 , wherein said decoding comprises MPEG-2 decoding of said merged video frames.