Tuesday, October 29, 2013

The Return of Photo Editing

So besides animation and visual effects, I've also been working on Photography, specifically Photoshop retouching.  Now I am very familiar with all the tools used in retouching in Photoshop, and I am familiar with many of the basic and advanced techniques used.  As such, I was worried I would be in a Photoshop class that would teach me nothing.  So far, while the class has been mostly review for me, I have learned a few of the more airbrushing techniques I haven't used before.  YAY, more to learn in Photoshop!

So right now, I have been learning portraiture retouching, working with multiple layers of edits and a variety of techniques that not only use a non-destructive workflow, but also provide flexibility to do edits I found tedious and difficult before.

So hopefully I will have 2 or 3 images to post soon.  I'll keep you all posted (I wanted to make sure everyone is alright with me posting them here before I do it).  And who knows, this could lead to some more freelance work or photography gigs.

Sunday, October 27, 2013

An Intro to Shader Writing

So lately, I've been learning how to do shader writing.  What does that mean?  Well for anyone who uses Maya (or any other 3D software), the shader is the thing that lets you apply a texture or color to your geometry, and interacts with the lights in the scene, such as spec highlights, diffuse, glow, incandescence, etc...  Lambert, Phong, Blinn, and Phong E are all examples of shaders.  What I am just starting to learn is how to write my own shaders.  Eventually, I can even set these up so you can plug in the variables for the Maya users to adjust, to make it user friendly.
Right now, it really just looks like a grid, well a weave at this point, which isn't like any other shader I just mentioned.  It has none of the light-interaction controls yet, and the other shaders don't control patterns like this one does.  However, this is just a learning tool.
It has been really fun to program so far, and there is still more to do on it, but I figured I'd post what I have so far and let people see.  I plan to add more colors, and maybe another pattern to this.  We'll see how it turns out.


In other news, I have been doing some serious Photoshop editing, but I'll go into that on another post.  The long and short of it is, I haven't really learned any new tools in Photoshop since I started, but I did learn some techniques I have never used before (which is what i was hoping to learn).  I kind of wish I could take their advanced Photoshop class to really dive into it.  We'll see.

Thursday, October 17, 2013

Snow globes

Just in time for Christmas.... oh wait, isn't there another Holiday or 2 between now and then?  Ah who cares about those holidays, we all know the only holiday that counts is the one where people spend the most money.  At least that's what stores seem to think.  Honestly I love Halloween and Thanksgiving, but for my latest project in my Renderman class (VSFX 319), we had to model snow globes.  While one girl in class actually did do a Halloween themed snow globe, I couldn't help but equate snow globe with Christmas.  So I decided to model Santa's Workshop.

The assignment was specifically to learn IBL (Image Based Lighting), which is a way of taking a very specific image (HDRI) of an environment and putting it in a scene, allowing the computer to calculate the lighting of the scene based on the values in the image.  It's kind of like taking an environment and putting it into your scene.  Sounds magical, but it is really not going to look right.  Imagine taking motion capture data and just leaving the raw data on the character and displaying that in a movie... it doesn't look good, trust me.  Animators need to adjust and tweak and fix the raw data into something that looks good and suitable for the character.  Just like that, I have to take this IBL and tweak it, add lights, adjust values, etc. to make the lighting look realistic like the object belongs in the scene.

We also had to use a specific Global Illumination technique, and actually got to play with 2 methods of it.  One uses Point Clouds to calculate illumination and occlusion, while the other uses ray tracing.  The point cloud based method is faster, stores the data to speed up processing later on, and developed by Pixar, but prone to inaccuracies.  The raytracing method is longer, but more accurate.  I used a hybrid technique where I raytraced, and then stored the data to speed up the processing.  It took about an hour and a bit to render these, but it was surprising to see some students with render times ranging from 20 minutes to over 4 hours!

I haven't quite dialed in the lighting all the way yet, and I am still working on the camera focal length (so expect more updates on this one later), but for now, I am relatively pleased with the results of my work, but I'm not going to leave it there.  For now, though, enjoy these three images.  Ho ho ho.




Saturday, October 5, 2013

SCAD Generate 2013

Hey everyone!

So I just finished a competition for art students here at SCAD Atlanta, called Generate.  It is a 24 hour challenge where each department gets sponsored by a different company and we have 24 hours to break into teams and create an original work based on their criteria.  They often give prizes, free food, and have a lot of fun.  We had trivia (which we were surprisingly bad at), which helped lighten the mood, and waaaay too many sugary foods.

All that aside, our team was part of the Visual Effects challenge.  But why is an animation major taking the VFX challenge you ask?  Well this year the VFX department added a Technical Direction challenge which was right up my alley.  Our goal was to spend 24 hours making an environment for a character.  We didn't need to include the character, but we did have to describe the character.  We had to include an element of animation and part of our project had to use Houdini in some aspect.  Our character was a woman named Fern who lived in 1945 in Paris, France.  She has been getting lettings from her husband, a soldier in the war, and the last letter promised to meet up with her at the cafe they first met at.  Fern waits for him at Cafe Maison, but he never shows.

Sad, I know, you can blame me, I came up with the sappy details.  And I did a ton of work on this with 6 other students, all contributing their skills to the project.  Nicole and I were the two Houdini artists, so we made a billowy smoke effect for a blown out candle on the table, but she really did all the work there.  I was mainly in charge of making sure the work got divvied out and shaded most of the glass work and such.  I was also the only one who had significant experience in Nuke, so I was given the challenge of compositing the project with literally minutes till the deadline.  I had a TON of fun, and we managed to finish a rather complex scene in JUST under the wire.

While I would love to say more, for now I will leave you with a single still image of one of our two shots.  I know it isn't perfect yet, but with a little more work (and a LOT more sleep), it will look amazing!  Enjoy!


Thursday, October 3, 2013

Readressing early work

I am interested in trying to re-work some of my earlier projects.  I was thinking it would be a great idea to remodel, texture, light, and render something I did toward the beginning of my time SCAD, and compare the two images to see how far I've come.  I haven't had the chance to do this yet, and I still have a class or two that I need to take to really make this look great, but I think it will make a great study.

The two images I am thinking about are from my introduction to 3D, "Forms, Space, and Lighting".  Right now all I can do is look at what I've done and think about what I want to do with it.  I think it may be a great segue into developing some more portfolio pieces on my own.

So here are my earlier works, an environment based on the character Harry Dresden, from the series of books The Dresden Files, by Jim Butcher.  For those unfamiliar with Dresden, he is an openly practicing wizard in modern day Chicago, who works as a private investigator.  It is adult, smart, well thought out, and entertaining.  Check out the books, and enjoy the images of his sub-basement lab.




Wednesday, October 2, 2013

Photography assignment 1

This week we went over our photography assignment in my fundamentals 2 class.  The class is really focusing on Photoshop, something I am incredibly familiar with already.  This first project actually didn't involve any Photoshop at all, but did go over Lightroom editing tools I wasn't familiar with, so I am glad to have learned them. 

My concept for this project was to explore the interplay of themes at the Davidson Arabia Natural Preserve, specifically the nature aspects versus the derelict man-made structures that still riddle the area.  For now, enjoy the images.











Teapots and Ribs. Yum!

This quarter I have been taking a Renderman class.  Well, it is technically PR Renderman (Photo Real Renderman), which is really just a production ready and incredibly powerful rendering tool.  It is used by a lot of large studios, and developed by (surprise) Pixar.

Right now, we are programming in scene description files, known as "ribs", which is like coding a scene by hand, geometry, lights and all.  It is actually rather challenging, but fun when you get used to it.  But it is slow, not fast at all, but knowing how the codes work is really useful.

Our latest assignment was to make a scene with a teapot, and some of the student's works were spectacular!  And yes, these were all HAND CODED (although we did get to use Cutter to help).  I decided to use this year's Renderman Walking Teapot toy I got from SIGGRAPH 2013 as reference, but the shapes were too complicated for me to figure out in the short time we had.  So I used it as inspiration, and made it my own.

The concept I had for this piece was to make it look like a toy tea set, since I used a toy as inspiration.  There is more I would like to do to this, but for right now, I am rather happy with what I have.  Feel free to leave feedback or comments!