A video coding apparatus carries out a compression coding of a video based on a result of predicting a motion vector predicted between frames of the video. A quantization parameter control unit comprised of the apparatus calculates a ratio of an accumulation value, in the small block, of a prediction error of a brightness component to an accumulation value, in the small block, of a prediction error of a color difference component at the time that a motion vector detection/compensation unit which carries out a motion compensation based on a prediction result of the motion vector for the small block designated in a frame of the above video as a prediction accuracy indicator of the foregoing motion vector, and carries out a control of a quantization unit executing a quantization of image information of the small block based on the prediction accuracy indicator.