Inter-layer prediction method for video signal

摘要

The present invention relates to a method for using interlaced video signal of a base layer in interlayer texture prediction. The present method separates interaced video signal of a base layer into even-field and odd-field components, interpolates the even-field and the odd-field components respectively in vertical and/or horizontal direction, and constructs a combined video data by interleaving the interpolated even-field and odd-field components.

主权项

1. A method for encoding or decoding a video signal with a decoding apparatus, the method comprising:
deriving, with the decoding apparatus, position information of a reference block in a base layer based on position information of a virtual reference block in a virtual base layer, the virtual reference block corresponding to the current block in an enhanced layer, wherein the reference block is a field block and the current block is a frame block; obtaining, with the decoding apparatus, texture prediction information of the current block in the enhanced layer when the reference block corresponding to the position information of the reference block is an intra-coded block; and obtaining, with the decoding apparatus, motion prediction information of the current block in the enhanced layer when the reference block corresponding to the position information of the reference block is an inter-coded block, wherein the motion prediction information of the current block in the enhanced layer includes a motion vector and a reference index, wherein obtaining the texture prediction information comprises:
a) constructing, with the decoding apparatus, a base layer picture in the base layer, the base layer picture including the reference block;b) applying, with the decoding apparatus, a de-blocking filter to the base layer picture;c) interpolating, with the decoding apparatus, each of even and odd field components of the de-blocking filtered base layer picture in a vertical direction and a horizontal direction;d) interleaving, with the decoding apparatus, the interpolated even and odd field components; ande) obtaining, with the decoding apparatus, the texture prediction information of the current block in the enhanced layer by using the interleaved even and odd field components, wherein obtaining the motion prediction information comprises: obtaining, with the decoding apparatus, a vertical component of the motion vector of the current block in the enhanced layer by multiplying a vertical component of a motion vector of the reference block by two; and obtaining, with the decoding apparatus, the reference index of the current block in the enhanced layer by multiplying a reference index of the reference block by two.