Provided is a stereo endoscope apparatus, including: a treatment instrument that is operable at least two imaging units for picking up images of a subject and a transparency-composition unit for performing, in a first region in which the treatment instrument appears within a first image picked up by one of the at least two imaging units, image composition by blending, at a predetermined composition ratio, an image of the first region and an image of a second region in which the treatment instrument does not appear within a second image picked up by another of the at least two imaging units.