Here’s the rowdy, relevant, relentless & rare rendering we rounded up for you this week. Enjoy!
Going to Siggraph? Lucky you! If you’re going then you should be there right now & probably aren’t reading this. But if you’re still en route or don’t have a plan, then check out these calendars of courses, papers and other events here
A Machine Learning Approach for Filtering Monte Carlo Noise (SIGGRAPH 2015) by Nima Khademi Kalantari
The video discusses a new method for handling render noise via machine learning/AI. The process is well diagrammed in the video and I can see this having huge potential, though I can’t imagine the processing power required. Please watch the video and let us know what you think about this process in the comments.
HT to Seghier Khaled for posting this to our Facebook page.
Counterpoint: CGI is not bad, you just don’t notice when it’s good by Joe Berkowitz
Joe takes his cue from a video posted on YouTube showing how good CGI is essential to creating the best scenes in the movies we love. You might immediately want to disagree, but I think there is a lot to this.
It’s easy for us to think of bad CGI like this:
But the great CGI effects in movies like Avatar, Terminator 2 (for it’s time), any of the LOTR movies or Mad Max: Fury Road doesn’t come to mind very quickly. Those movies worked so well that the CGI never stands out, and so we praise the movie instead of praising the effects. That is the point of the article by Joe and the video essay that he cribs the idea from.
I cited the Fast Company piece here because that’s where I learned about the video essay. But here is the actual video essay:
and the video its (sort of) replying to:
That’s it for this week. Thanks for dropping.