Photogrammetry and Stuff

I read Andrzej Poznanski’s article; ‘Visual Revolution of The Vanishing of Ethan Carter’, and was quite fascinated with what they could do. However, I am wondering, getting the model to render appropriately was quite some trouble, the main pro of this method was to get believable wear and tear etc, and the model must still modified to reduce poly count and get a nice edge flow. Would it not be just a easy to just take reference photos, model the model yourself, then only take photos for textures? I have not yet modeled any high-res models, nor do I have a lot of experience with photogrammetry, and they would not be using it if it did not pay off, so I am obviously missing something.

As a kid, I used to do a lot of Warhammer painting, as well as terrain making. Crafting model miniatures is something that I have always been interested in, and  while making scans of environments from real life is of moderate interest to me, something that I would really like to try out some day is making a game composed of actual hand crafted props. Imagine for instance a Harryhausen or Burton inspired game that actually looks like a stop-motion movie, or a maybe something in fashion with ‘The Dark Crystal’. In short, I think this tool can be (and probably is) used for more than just realistic graphics. If miniature sculpting for games was a profession, it would probably be my dream job.

A pro for using photogrammetry for miniatures instead of outdoor environments, is that you would not have to worry about the weather. Of course, since you have to sculpt the model yourself, the whole point of using photogrammetry to capture the creations of nature is lost.

When modeling faces, one thing that I find important to think about is that many of the loops circle around certain features, mainly mouth and eyes, rather than the entire head. This seem to be mainly due to that the face muscles actually circle these features in that way, and it would look right if lines were drawn differently. However, it also saves polygons, since we need a high level of detail in the face, relative to the back of the head, and since the loops for the details around the eyes and mouth can be restricted to that area, we do not have to waste polygons circling the entire head.

The nose/cheeks and forehead are the areas that have horizontal loops circling the entire head, they are really just part of the base shape of the head (except for the nose being slightly extrude) and really do not have that much detail. The nose tip and nostrils probably make up the most interesting topology in the face. They nostrils have individual loops such as the mouth and eyes, however, the between the tip and the nostrils, the loops the there are also some of the outer loops of the mouth.