| 摘要 |
<p>The invention relates to a method of video encoding and decoding, on the basis of a reference image (F1, F2) and of a field of motion vectors (M1, M2), there is calculated a predicted image (P2, P3) and a residual image (R2, R3). During encoding, the method comprises a step of generating a list of indicators of order of occlusion which is transmitted to the decoding with the fields of motion vectors (M1, M2), the predicted image (P2, P3) containing occlusion bands formed by pixels whose vector field does not match a valid position in the reference image (F1, F2). The list of indicators of mode of filling of pixels of the occlusion bands of the predicted image (P2, P3) is transmitted from the encoding to the decoding, according to which list, during decoding, there is tracked a pre-established mode of filling so as to fill each pixel in said occlusion bands of the predicted image (P2, P3).</p> |
