A video decoding apparatus using an intra prediction, includes: a decoder to decode a bitstream to thereby extract residual signals; an inverse quantizer to inversely quantize the residual signals; an inverse transformer to inversely transform the inversely quantized residual signals; an intra predictor to predict a current block from a plurality of neighboring blocks of the current block; and an adder to add the inversely transformed residual signals to the predicted pixels in the predicted block, wherein the intra predictor includes a weighted pixel calculator to calculate a plurality of weighted pixel values of pixels selected in the neighboring blocks by applying weights to the pixels selected in the neighboring blocks, the weights being decided variably based on a position of a current pixel to be predicted in the current block; and a pixel predictor to derive a predicted pixel value of the current pixel using the weighted pixel values.