| 摘要 |
The present invention concerns a method and device (10, 20) for the processing, in particular coding or decoding or more generally compression or decompression, of a video sequence (110) consisting of a series of digital images (401 to 407) comprising a current image (401) to process. The method comprises the steps consisting of: —generating (511, 603, 720, 812) a first reconstruction of a first image (402 to 404) in the sequence, so as to obtain a first reference image (517, 610, 717, 810); —generating (519, 612, 720, 812) a second reconstruction of the first image (402 to 404) different from the first reconstruction, so as to obtain a second reference image (518, 611, 718, 811); and—predicting (505, 606, 705, 806) at least part of said current image (401) from at least one reference image selected from the generated first and second reference images (516, 608, 716, 808). |
| 主权项 |
1. A method of processing a video sequence in which the video sequence comprises a series of digital images including a) a current image and b) a first image prior to the current image, the method comprising:
1) generating a first reconstruction of the first image and a second reconstruction of the first image, the second reconstruction being different from the first reconstruction, wherein the first reconstruction is generated so as to obtain a first reference image, and wherein the second reconstruction is generated so as to obtain a second reference image; 2) selecting at least one reference image from the first reference image and the second reference image; and 3) predicting at least part of the current image from the selected at least one reference image, wherein the images are defined by a plurality of data blocks, and wherein the second reconstruction of the first image comprises a reconstruction solely of blocks of the first image that are used during a prediction of at least part of the current image. |
