Provided is an interlayer video decoding method according to an embodiment including obtaining prediction-mode information of a current block of a depth image from a bitstream, generating a prediction block of the current block on the basis of the prediction-mode information, and decoding the depth image by using the prediction block. The obtaining of the prediction-mode information of the current block from the bitstream includes receiving a first flag, a second flag, and a third flag, wherein the first flag indicates whether prediction of the current block by dividing the current block into two or more partitions according to a pattern is permitted, the second flag indicates whether the depth image permits blocks of the depth image to be predicted by dividing the blocks into two or more partitions by using a wedgelet, and the third flag indicates whether the depth image permits the blocks of the depth image to be predicted by dividing the blocks into two or more partitions by using a contour; and receiving a fourth flag from the bitstream when predetermined conditions determined on the basis of the first to third flags are satisfied, wherein the fourth flag represents information regarding a type of a method of dividing the current block into two or more partitions according to the pattern.