DSN 6400 (DAIM Graduate Studio):
Approach: Memory Moments is due! The project that I have been working on for the past four weeks finally had the deadline approached. The final steps that I needed to approach were rendering and post lighting (in Premiere) before Tuesday's (Feb. 5th) class.
Choices Made: For the few days (Sunday and Monday) leading up to the final deadline for Tuesday, I decided to work backwards in renders according to the shots. I wanted to render out shot 3 first since it had the heaviest lighting and longest shot - the one I found most important to the project. It also was the longest shot in terms of frames (500). So I rendered out that shot and did post work on it - each frame taking roughly 4~5 minutes to render.
From there I went to the bicycle shot (shot 2) and had to make adjustments within there for background trees. Since I decided to use Arnold as the renderer, and the HDRI map I used for the skydome wasn't showing any leveled plane, I needed to create more of the environmental trees in the back, and light the shot accordingly so that it didn't look like the ground was flat and falling off the edge of the Earth. So I populated the background with more of a forest look to it, and then created moss and brush on the ground to disguise the forest floor. From there, the lighting fell into place with spotlights creating highlights along the ground, and the bike riding past the trees. This render had only 60 frames to render, so that didn't take too long, but the post lighting took a bit more effort with the bike motion cutting across the scenery.
The last shot I focused on was the underneath view of the Spanish moss along a straight path. This lighting was by far the absolute hardest because I really wanted to get 'godrays' in my shot, but I didn't learn how to do that in Arnold until after I had already rendered. So I probably made about ~15 spotlights to cast down through the moss to light it to begin with, and then have highlights along them. But that was easier said than done. When I couldn't achieve the look that I wanted, I simply relied on what I could do in premiere (enhancing brightness, exposure, contrast, overlaying godray images) to somewhat get that sunbeam effect in this one, while still having the camera move during the shot.
Next Steps: During the review session, Maria made a comment on my second shot about the textures being off from the rest of the scene. It's like I have two different asthetics happening - and it is very obvious that the bike/girl textures aren't matching the environment that I have placed them in. So I would like (in my free time...which there is very little of as a grad student haha!) try to re-texture the bike and the girl and try to get them to match the scene. Then re-render them and get them edited in post to match the work.
Process Video of how I went about making the three scenes:
Other: This week also consisted of finalized proposals for our Phase I project for the next 5 weeks. For Phase I of our project proposals, I will focus on the technical problems my side of the thesis project has been having, or those issues that haven’t been given enough attention – the emotive states of the bodies and facial features of Lucille Bridges and the Government Official within the car scene. I have a lot of motion capture data to work with, clean up, and polish for proper staging, fluidity, and purpose. By this I mean, right now those (digital) characters are not performing in the style I am aiming for. Compared to the original capture and footage of the live performance, and mirrored into the digital performance, it can be pushed for more emotion, better body language, and period acting. I need to spend time working in MotionBuilder and Maya to tweak both body motion and facial motion, given that these characters have very different performance styles.
- Lucille Bridges: woman; silent; primarily focusing on body language; back seat placement
- Government official: male; lots of voice acting; primarily focusing on facial emotions; front seat placement
I also want to conduct research in some of the case studies I’ve listed below. Researching and analyzing the design choices in performance for these VR projects will help influence my own path in this project. So setting aside time for that study will be helpful.
ACCAD 5194.01 (Performance and Narrative in VR):
FILM 7001 (Feminist Film Theory):
Hennefeld, Specters of Slapstick & Silent Film Comediennes
GRA (Graduate Research Assistant):