Abstract:
The present invention relates to a texture description method in frequency domains and a texture-based retrieval method using the same, in which image signals are converted into those in a frequency domain of Polar coordinate system. The texture description method in the frequency domain comprises: a first step of generating a frequency layout by partitioning the frequency domain into a set of feature channels; a second step of extracting the texture feature values of the image from the respective feature channels; and a third step of constituting a texture descriptor of the image in a vector form using the texture feature values extracted from the respective feature channels of the frequency layout. According to the present invention, in a texture-based retrieval described in the frequency domain, a retrieval method capable of satisfying respective or all the retrieval conditions such as scale-, rotation- and intensity-invariant retrieval of relevant texture images is provided. Also, a retrieval method of computing the rough scheme of texture to provide a browsing functionality in retrieving is provided.
Abstract:
Disclosed are an image encoding/decoding method and device supporting a plurality of layers. The image decoding method supporting the plurality of layers comprises the steps receiving a bitstream comprising the plurality of layers; and decoding the bitstream so as to acquire maximum number information about sublayers with respect to each of the plurality of layers.
Abstract:
A method and a device for encoding/decoding an image are disclosed. The method for decoding an image comprises the steps of: decoding information on a quantization matrix; and restoring the quantization matrix on the basis of the information on the quantization matrix, wherein the information on the quantization matrix includes information indicating a DC value of the quantization matrix and/or information indicating differential values of quantization matrix coefficients.
Abstract:
An inter prediction method according to the present invention comprises the steps of: selecting candidate units from among reconstructed neighbor units; creating a candidate unit set with respect to units to be decoded, using the selected candidate units; determining a reference unit from among the candidate units which constitute the created candidate unit set; and performing inter prediction on the units to be decoded, using the determined reference unit. According to the present invention, image encoding/ decoding efficiency can be improved.
Abstract:
The present invention relates to a method and apparatus for encoding/decoding image using a motion vector. According to one embodiment of the present invention, an image-decoding method is provided. The image-decoding method comprises the following steps: clipping a motion vector of a reference picture with a predetermined dynamic range so as to generate a clipped motion vector; storing the clipped motion vector in a buffer; deriving a motion vector of a block to be decoded using the motion vector stored in the buffer; and performing inter-prediction decoding using the motion vector of the block to be decoded. According to the present invention, the size of a memory space required to store a motion vector may be reduced.
Abstract:
According to the present invention, an image encoding/decoding method comprises the steps of: performing an intra prediction on a current block so as to generate a prediction block; performing filtering on a filtering target pixel in the prediction block on the basis of the intra prediction mode of the current block so as to generate a final prediction block; and generating a reconstructed block on the basis of a reconstructed differential block corresponding to the current block and on the final prediction block. According to the present invention, image encoding/decoding efficiency can be improved.