1. An image processing device comprising:
circuitry configured to operate as:
a predictive motion vector generating unit that:
sets a co-located vector to be available and to be scaled in a candidate list as a condition that both a type of a reference picture of a current block and a type of a reference picture of a co-located block are short-term reference pictures;
sets the co-located vector to be available and to be not scaled in the candidate list as a condition that both the type of the reference picture of the current block and the type of the reference picture of the co-located block are long-term reference pictures;
generates a predictive motion vector of a coding vector of the current block using the candidate list; and
a reconstructing unit that reconstructs the coding vector by using the predictive motion vector generated by the predictive motion vector generating unit; and
a decoding unit that decodes the current block using the coding vector reconstructed by the reconstructing unit.
2. The image processing device according to claim 1,
wherein
the predictive motion vector generating unit scales the co-located vector based on a positional relationship between a current picture and the reference picture

referred to generate the predictive motion vector of the coding vector.
3. The image processing device according to claim 1,
wherein
the circuitry comprises a Central Processing Unit (CPU).
4. The image processing device according to claim 1,
wherein
the reconstructing unit uses a merge mode to reconstruct the coding vector.
5. The image processing device according to claim 1,
wherein
the reconstructing unit uses an AMVP mode to reconstruct the coding vector.
6. An image processing method performed by an image
processing device, comprising:
setting a co-located vector to be available and to be scaled in a candidate list as a condition that both a type of a reference picture of a current block and a type of a reference picture of a co-located block are short-term reference pictures;
setting the co-located vector to be available and to be not scaled in the candidate list as a condition that both the type of the reference picture of the current block and the type of the reference picture of the co-located block are long-term reference pictures;

generating a predictive motion vector of a coding vector of the current block using the candidate list;
reconstructing the coding vector by using the generated predictive motion vector; and
decoding the current block using the reconstructed coding vector.