Hello Everyone! I Make Modern Filmmaking Tools For YOU!

ok, sure, can you explain the basics of your technology's functionality? I'm well versed in modern filmmaking tech, and working on a system of my own to expand the art.

Is this a vr/ar format?

is this a calibrated multicam setup?

Are you using virtual characters?

What is special and unique about your system? Anybody can make AR/VR characters and create a film now, or do a 360 environment scan. What is the new aspect of this technology that merits a new name?
 
I'm saving my pennies for a capture suit. The kind that uses accelerators and gyros and a pinch of hocus pocus. $2,500 and you are in business. Michael, are you guys still in the R&D phase or have you got a working system? How about a website?

The thing with volumetric captures is occlusion, which multiplies if you plan on capturing more than one person at a time, unless you're using some sort of AI to predict movement based on known rotation point in a human skeleton.....

I have a silly little mocap setup that has a maximum volume around 20'x20' with 5 or 6 cameras but I'm happy with it. It suits my needs but I'm always interested in learning about new approaches to the whole mocap thing..
 
Last edited:
I'm saving my pennies for a capture suit. The kind that uses accelerators and gyros and a pinch of hocus pocus. $2,500 and you are in business. Michael, are you guys still in the R&D phase or have you got a working system? How about a website?
Just FYI, if you're looking to get into mocap on a budget, they now have tech that can, to some degree, capture mocap data via video analysis.


A good mocap suit is around 12 grand, but you can get decent results from the less expensive ones, and there are kind of a lot of alternatives popping up this year for this part of the pipeline.

This AI mocap, or video driven mocap, is available from a lot of sources now, with everyone from Nvidia down to Iclone jumping on the bandwagon. UE can drive a metahuman directly from your iphone camera in just a few steps for example.

Here's a quick test I did the first time I tried using it. I just did some facial expressions, to test the tech. It's not good, but I managed to recreate myself as a dead eyed factory worker. lol.

 
Last edited:
Cool stuff.
Yes, mocap is available to the masses. Now the most important question that nobody really things about; 'what shall I do with it?' My advice is to not do humans. Almost nobody but WETA and sometimes ILM get it right. Nothing worse than humans that look like articulated manikins driven by under powered motors from Radio Shack.
 
You can make humans look good, but it's very difficult, and takes significant time and money. We may be using the same 15k mocap suit as Lucasfilm (rip) but what we don't have is a custom warehouse, huge lighting rigs, and actors that are in a great, energetic mood because they are getting paid one mortgage per hour. People just turn in a different caliber of performance when they know that the work is getting wide release, and also studios have post guys like us go back and fine tune the mocap data to smooth over the rough edges, making some of their products look better than any raw output from a capture ever could.

Tech is a huge deal, obviously, but always remember, it's more about the player than the guitar.
 
To a certain extent I disagree. Performance is very important and accurate mocap is next in line, but, what does on under that polygonal character mesh is the unsung hero of truly believable CGI humans. The custom scripts, mind boggling rigs, blend shapes, clusters, muscle systems,,, not to mention the multiple layers of animation that is keyed over the mocap by skilled animators. This is not kids stuff. Talented people outside of the studio environment can still hope to do animals and dinosaurs and monsters and fish and birds, but humans? no way. Too difficult to fool the audience.... and the reaction from audience members when a CGI human is not spot on.. I mean, there is no forgiveness. In the days of Dick Smith and Rob Bottin, prosthetic make-ups were admired and appreciated even though you could always sort of tell they were rubber glued onto the actor's face. They looked good but they were rarely convincing and they never really fooled anyone. CGI is not afforded that much leeway. If it isn't perfect, we HATE it. I'm not talking about industry people. I mean the average person watching a movie has a low tolerance for not so good CGI.
 
Last edited:
I definitely know what you mean. The area where this fluxuates is in the demographics of the viewership. Younger audiences tend to be more forgiving. Stylizing avatars is a quick fix that works. Obviously Frozen 2 isn't going broke, but then the humans aren't portrayed realistically. The uncanny valley has haunted would be cgi filmmakers for a generation, but we now have many cases where the issue has been effectively circumvented.

I basically agree with what you are saying, but here's an interesting counter question. We all know why realistic characters fall flat with audiences, but can you explain why much cruder representations of humans fare just fine. Peter from Family Guy, or Shaggy from Scooby Doo don't look much like humans, but our minds accept them much more easily than for example the notorious characters from FF Spirits Within.

Why are we all perfectly ok with a 2/10 accuracy representation, but annoyed and disinterested in the 9/10 version?

In my test video, you can see every individual hair on my head, skin pores, beard stubble, but it's worth 0 cents. Scooby Doo was translated into 15 languages and made more money than most literal gold mines. And what about Ziggy? Why was Ziggy a more popular character than for example the Navi from Avatar?

1656381024227.png
 
Last edited:
Is this guy ever going to come back and tell us what his tech is? I don't even have a company name to search. Volumetric could kind of mean anything, is this voxel based? How does the cinematography aspect fare in this system? If it's user view based, is this just a machinima system for AR/VR? What's the install base on your client end? Is this live action volumetric?

Can you provide a sample film made using your tech?
 

indietalk

IndieTalk Founder
Staff Member
Admin
:welcome:
 
I basically agree with what you are saying, but here's an interesting counter question. We all know why realistic characters fall flat with audiences, but can you explain why much cruder representations of humans fare just fine. Peter from Family Guy, or Shaggy from Scooby Doo don't look much like humans, but our minds accept them much more easily than for example the notorious characters from FF Spirits Within.
Apples and oranges. Cel animation is what it is. It was never ever trying to pass itself off as photo realistic. People love Bugs Bunny but nobody is upset that he is painted on clear sheets of acetate. Cel animation and comic books. The same thing except one is animated and one isn't We love them but we don't expect them to be 3D and photorealistic.

If you want to put a 9/10 cgi human in a movie, the movie better be akin to TRON. Something with a storyline that might explain why the people are not quite people looking..
 
In memoriam - Michael B Chapman

I think we all remember the magical time in our lives when we first met Michael, an energetic and loquacious personality, who first introduced us to the expansive world of information surrounding volumetric filmmaking. Who can forget the early part of Michael's life here on the forum, when he shared his name and vocation with the community. But as special as those times were to me, I think it was the middle part of our relationship that left such a deep and inspiring footprint on the way I think about art. His tireless efforts to enlighten the world about a new way to make film where we could all escape the surly bonds of cameras and greenscreens and soar into a new era of creative composition unfettered by conventional obstacles was an inspiration not only to me, but to a generation.

Unfortunately, after repeated attempts to revive him, he was recently deemed "unresponsive". As sad as this occasion is, I think it's important for us to remember the good times, and not dwell on the negative circumstances surrounding his untimely departure.

Micheal "ask me anything" Chapman - june 27 2022 - the afternoon of june 27 2022
 
Last edited:
A great man. He and I became very close, but alas, a man who dances to his own tune doesn't stick around very long and so he left, never to be seen again.
 
Top