Thursday, October 13, 2011

Shorter Week

This week I didn't work far enough on Senior Project as I wished. Fall Break was this weekend, which meant it was time for me to play some catch up with the readings in one of my classes and try to finish up my other programming assignments (B-Splines == nuisance). So I didn't work on my project until Monday. What I did do was this:

I managed to pull the MEL script for the automatic rigging into Maya. It took a while to figure out because the instructions weren't clear, but I've been able to test it on a model, and it's pretty cool. I'll post a video later this week, I want to iron out some kinks I have from not having all of the scripts in place. The model I've used is the basic model, an OBJ file, that I got from Pinocchio. While this turned out fine, it doesn't really help me see where I really need to fine tune the rigging, so I also want to see how well the auto-rig tool works on a more realistic model. I did a lot of model searching, trying to see if I could find free models that I could pull into Maya of different shapes and sizes. Unfortunately, that's proving to be a little difficult and tedious at the moment. A lot of the models I found don't come in a compatible file format, I did find this one model, which I will use as a test for now!

So much more realistic!

Anyway, besides the rigging, I'm in the process of gathering my data. I'm renting a camera from the library this weekend, and I'm going to record some people walking around campus. Hopefully by the end of this weekend I'll have some data to start off with!

Also, I've been looking through Youtube for helpful videos. I really like this video because they slow it down for you, and talk you through how a walk is supposed to look. The video is kind of hypnotising, actually. 



Youtube is turning out to be a really great resource for seeing some focus on the movement itself. While my videoing will be useful, I like that I can find videos of people just walking on a treadmill for you with only their legs in focus if I just want to get an idea of the simple, overarching movements.

Lastly, my main goal for the next week is to script a walk-cycle, so I can have something that combines everything I've done so far together. I'm really excited for this, it'll be the beginning of things to come. From the scripting, then I'll be able to focus on the algorithms and how they affect my walk-scripts or whether my algorithms create a walk-cycle from scratch each time. I'm not sure which approach I'll take, only time will tell.

4 comments:

  1. I really would love to see some videos of this.
    I can send you some motion capture ROMS to test out the rigs this generated.

    ReplyDelete
  2. Treadmills are bad! FYI.
    Though they give some idea of how humans are rigged they are NOT NOT NOT good examples of walking or running. Norm will go off on a 20 min rant, lol. People lean differently when on a treadmill as they walk and run. Though its a good idea to see how feet and ankles move. They are NOT good examples of 'true' walking and running.

    ReplyDelete
  3. Also can you post detailed images of the rigs and weights the tool generated?

    ReplyDelete
  4. Yeah, I planned on using the treadmill videos more for the rig, because this shows the joints in focus and how they can move around, but I was planning on using my videos of people walking as my primary data.

    ReplyDelete