ENCODING DEVICE AND ENCODING METHOD, AND DECODING DEVICE AND DECODING METHOD

摘要

The present technique relates to an encoding device and an encoding method, and a decoding device and a decoding method that can improve encoding efficiency when a motion compensation operation with fractional precision is performed upon inter prediction. When the precision of a motion vector is ¼-pixel precision and the precision of a predicted vector is ⅛-pixel precision according to detected-precision information contained in compressed image information, a predicted-vector transform unit performs a rounding operation on the predicted vector to generate a predicted vector with ¼-pixel precision. A motion vector generation unit adds the predicted vector with ¼-pixel precision to a difference vector contained in the compressed image information, to generate a motion vector. An inter prediction unit and an arithmetic operation unit decode an image by performing a motion compensation operation using the motion vector. The present technique can be applied to a decoding device, for example.

主权项

1. A decoding device comprising:
a reception unit that receives an encoded image, a difference between a motion vector of the image in an inter prediction and a predicted vector, and detected-precision information indicating that precision of the motion vector for when a prediction direction of the inter prediction is bidirectional is lower than precision of the motion vector for when the prediction direction is unidirectional, the predicted vector being a motion vector of an image located close to the image; a predicted-vector transform unit that performs, when the precision of the motion vector is lower than the predetermined precision and the precision of the predicted vector is the predetermined precision according to the detected-precision information received by the reception unit, a rounding operation on the predicted vector to generate a predicted vector with a lower precision than the predetermined precision; a motion vector generation unit that adds the predicted vector with a lower precision than the predetermined precision generated by the predicted-vector transform unit to the difference received by the reception unit, to generate a motion vector; and a decoding unit that decodes the image by performing a motion compensation operation using the motion vector generated by the motion vector generation unit.