Spatial Computing Part II: Shopping

Part II:

[youtube][/youtube]Okay, admittedly this took way longer than I had originally expected.  I'm pretty happy with the results, and have been getting some decent feedback - so cool!

Also, I'd like to take this space to acknowledge everything out there that made this project possible:

  • Blender, the blender community, and especially super3boy's blender tutorials.
  • Google Sketchup and Google 3d Warehouse and all the folks who were awesome enough to make all those 3d models.
  • and all the folks there who were awesome enough to make all those sounds I used.

& here's a list of sources.


So some people have been asking me how I did it - if I used AR toolkit or whatever.  This project was almost exclusively made with Premiere and Blender, with lots of help from Google Sketchup and After Effects.

The workflow was this

  1. Storyboard and script the concepts that I wanted to explain
  2. Measure out the irl room (measuring tape, paper)
  3. Place the camera in the room and measure it's location (measuring tape, paper)
  4. Record footage of me talking in an empty room
  5. Model the room and cameras in Blender
  6. Place the digital camera in the model of the room (Blender)
  7. Build out the 3d models that I wanted in the room (Blender)
  8. Paint the walls of the digital room Bright green (Blender)
  9. Export render from Blender
  10. Import into Premiere and greenscreen the footage.
  11. Superimpose my body (for shots where I occlude the footage) on top of the Blender footage
  12. Add effects, like alpha glows and the line that pops outa my finger (that was done in After Effects)
  13. Add Sound
  14. ???
  15. Profit!

Lastly - Why am I doing this?

*hoists self up on soap box*

Here's the deal - AR is awesome.  I feel like I'm preaching to the choir when I say that someday in the not too distant future AR will be in our lives.  But what are those next steps?  How do we begin to introduce AR consumer electronics into the mainstream market?

The way I see it, the biggest problem isn't that we lack the ingenuity or the engineering resources to make this happen.  The biggest problem is that from a mainstream perspective, AR isn't fun yet, it's still just an engineering experiment.

The best way to get AR consumer electronics into the mainstream market is to make AR approachable and exciting.  It's to convince the everyday person that AR is not just an experiment, but a tool that could actually be useful to them in their daily lives.  Something that could excite their perfectly natural and instinctive desires to physically stand in the same room with things (in this case, digital content).

So I decided to use the term "spatial computing" over "augmented reality" because I believe it is a more approachable term.  From an outsider's point of view, "augmented reality" almost suggests that reality is subpar or otherwise inadequate, "spatial computing," on the other hand, simply describes a computer intelligent enough to communicate spatially.

And that's why I'm building these videos - because I want to get people excited about what's soon to come.  I want to generate interest from the ground up.  For computing paradigm shifts this big (the computer mouse has been around for over 50 years!), we're going to need to swing a lot of cultural momentum.

**Update 2**

Actually, I take it back... sort of.

As I discovered in while writing my reply to this post over at augmx, the reason I use the term "spatial computing" is because it's actually different from "augmented reality."  In short, spatial computing is the activity of using a computer with 3d space as your main point of interface, while augmented reality is the layering of digital information over real world things.  Seriously, go read my reply to that post over there if you want to geek out with me over semantics.

The relative approachability of the terms, I believe, has to do with the approachability of the ideas they identify.  Almost everybody has experience computing, and absolutely everybody has experience with space.  It stands to reason (I'd hope) that these people would be able to pretty clearly and immediately understand what spatial computing means.