What does performance mean when assessing motion capture gloves? Here at StretchSense, we are relentlessly focused on making the world’s highest-performing motion capture gloves.
But I realized we haven’t actually talked much about what performance actually means, why it matters so much and what we’re doing differently about it.
If you look at a human hand, there’s an incredible variety of different movements that it can make.
I can spread my fingers, I can bend them, my thumb and finger can interact, my fingers can interact, my palm can collapse in interesting ways, my knuckles can move… and that’s before we even go into picking up objects, holding props, using tools and what have you. So, hands are very complicated.
In a high-performing motion capture system, what it will do is it will capture the highest number of those different complicated movements. If you think about it, maybe a hand can move a hundred different unique little subtle ways. A high-performing motion capture system will capture, say, all hundred, and then deploy them directly, without the need for cleanup, into final animated content.
In a low-performing system, you’ll try to capture those same movements and it won’t deploy them directly into final content. The quality will be low, you’ll need to clean up, you’ll need to spend time, money, and effort doing that.
So why does this matter to a studio?
There are really 4 key things that really matter.
1. More time
So, the first is time. If you have a low-performing finger mocap solution, you will have to pay your animators to sit there, sometimes for weeks or months, cleaning up bad, low-quality data.
Instead of seeing something the day after you’ve shot it or even the same day, you potentially have to wait weeks or months.
This slows down the creative cycle of the studio, it pushes out deadlines, and reduces the amount of work the studio can produce.
So, the biggest advantage of high-performing motion capture gloves or finger tracking solutions is time-saving.
2. Freedom to be creative
The second big advantage is that if you don’t have a high-performing system, and your animators are working on cleanup, they’re not really being maximally creative.
Animators didn’t go to school to clean up messy data from bad mocap performances. They actually want to create new and novel IP.
Having high-performing mocap data frees up time for animators to do things that are adding value.
It really sucks to go and spend the whole day doing a motion capture shoot, spending time getting people in, setting them up, running the whole thing, only to have no idea if you’re going to get garbage data at the end.
We could all live with that stress, but that’s not really fun. It’s much better to have a high-performing system and you actually can have confidence that what you put in is what you’re going to get out.
Obviously, paying people to clean up bad motion capture data is expensive, it costs studios a lot of money. We really wanna minimize that.
So here at StretchSense, we have a novel, new-to-market type of stretchable sensor technology. These sensors are wonderful for measuring fingers and unlike legacy sensor systems, they don’t suffer from occlusion, drift nor do they impede the motion of an actor’s hand.
With that strong, robust, accurate, comfortable sensor foundation, we can use machine learning in a way that no one else in the market can.
So when we have a new sensor design, a new glove design, a new piece of machine learning software, we put the gloves on the test group of people. We watch them move their hands around to see how many of those key poses or movements the glove captures with a high enough fidelity that it can be used directly in final animated content.
Our gloves get better over time as we add more data to them, and they can learn and be customized to the output that an individual studio actually wants.
Putting all this together, we’re proud of the progress that we’ve made in our quest to make the perfect pair of motion capture gloves.
And we’re so proud of it, we’re very comfortable to have people try them out.