A 3D image sensing device includes a first camera lens, a second camera lens and a light source. A 3D image processing method for the 3D image sensing device includes the following steps. Firstly, a target is photographed by the first camera lens and the second camera lens, and the obtained images are processed in a stereo vision mode. Consequently, a first depth map is obtained. After the light source emits plural feature points to the target, the target is photographed by the first camera lens and the second camera lens, and the obtained images are processed in an active stereo vision mode. Consequently, a second depth map is obtained. The first depth map and the second depth map are synthesized as a synthesized depth map according to a synthetization strategy.