Now, I must admit: I am very excited about 3D modelling. As a person who lvoes 3D printing, I was very excited this week to dip a toe into the process of scanning a real object, making it digital, and then setting it up with the potential to be 3D printed! My dreams can become a reality! As a musicologist, there aren’t a lot of ways I can immediately find this useful, as 3D printed instruments tend to work questionably at best and I don’t know of any other solid objects closely related to musicology. For art history, however, I can see so many ways in which this could be useful! As a method of complicating discussions surrounding aesthetics and historical reproduction, 3D scanning and modeling can certainly act as a research tool.
Briefly, Though, I would like to describe my experience using scanning technology. The program itself is fairly intuitive, which is something that pleasantly surprised me. With our former work on networks showing us some complicated software interfaces, my hopes weren’t high. But, as the software frays out sections you don’t need at the beginning, I found that very helpful. I’m easily overwhelmed by tons of options ad buttons with no clear usage, so this was intuitive for me. Uploading the images of the side of Person Hall on the UNC campus was straightforward, and the software walks you through uploading th eimages, creating preliminary dots, and expanding that to more dots anbd mesh, and finally creating the 3D model in rudimentary and later, more advanced detail. Even putting the texture back on the model was relatively easy! Most of these steps took just a few clicks. I’m very pleased with the finished product. Once a project is finished, it can be exported to PDF for presentational use, which I also found helpful.
The Agisoft Photoscan software worked alright for my own personal project, but not as well as the demo we did in class. I had a really difficult time figuring out how to get it to recognize all of the photos even though there was overlap. I sectioned out parts and removed them from the picture to eliminate confusing shadows that might mess with the software, but did not have much luck. Regardless of if I had an all-white or all-black background, the program still would not recognize about half of my pictures, at best. I spent a while trying to figure out these issues, to no avail. Below is the best that I could do with the antique pencil sharpener I used as my object. I followed the directions as best as I could, and still wound up with a sharpener that looks more like shrapnel.
- No wide or fisheye lenses, JPG is fine
- Send photos in orginial resolution to your phone
- Avoid smooth, shiny, mirror, or transparent objects – hair too
- Avoid unwanted foregrounds
- Avoid moving objects within the scene
- Avoid flat objects or scenes. Don’t edit the photos
- Take as many pictures as you can. Try doing entire circle high, straight on, and low
- Object should take up entire area possible
- Overcast days are great for this – if you can’t get soft light, get golden hour. No flash
- Keep camera straight on for the object, which means you do the moving
I’m interested to learn the next steps that take it from 3D model to printed object – from what I understand, there are quite a ew more steps to get it to that stage in meshlab or another software. That in and of itself is daunting, but I’m very excited and up to the challenge. My biggest concern for 3D modeling is understanding the implications of 3D images as research and pedagogical tools. I’m so excited to use and create with this software, and I feel like my need for creativity is making me so excited that it’s crowding my ability to be critical of 3D modeling on a higher level. I think that will come with more time and experience!