I’m working on a project that involves my actor holding a photo realistic object created in Blender, but I’m not sure how I would composite the model into the real footage, in which it needs to be placed directly in her hand (which forms a cupped shape). Does the actor need to hold an object-based green screen so it can later be replaced by the 3D model? Is rotoscoping involved? If someone could help me out with the process, that’d be awesome!
Probably at least an object track, green object would probably help to make the right hand gesture and help masking the object from hand/fingers. Could use a chrome ball rig to capture the set lighting, or a christmas ornament as a poor man’s solution
Place something in actors hand, I would personally not use green but a neutral gray object that is smaller than the 3D object but approximates the shape. Add tracking markers to it and do an object track. To get hand overlap, roto the hand/fingers that cover the inserted object.
As the actual object is not exactly the shape of 3D object, you would have to do roto even if using green object, so I would not bother with keying and despilling at all but roto everything. Much cleaner solution imho.