Launching ImageMatching beta

In our “crusade” to make Augmented Reality and Computer Vision available for everybody, we are proud today to launch our second product beta, the Image Matching.

It has passed one month since we published our first product, the ARBrowser and today we take the chance to release the beta 0.2 with some improvements and new features.

Any help, comment or inquiry received from the community will be extremely welcomed as our aim is to build tools from developers to developers. It is our scope to try to add all the requested functionalities that cover your targets.

 Image 1. An illustrated example of the ARBrowser.

During this month we have meet and received feedback from some developers that helped us to push and improve the engine so it would eventually cover their expectations. Indeed we specially had a lot of feedback from Todd Werelius, CTO – Founder at bThere, who actually helped us and gave us the opportunity  to ask him some questions:


Pablo: Tell us a little bit about yourself?


I have been doing commercial software development work for 20+ years now, I wrote an internet browser for Oracle before anyone thought much about the internet, a virtual reality browser that no one cared about, and a PPP (point to point) protocol stack for phones before anyone thought they would get “smart”.  If you want to know what to develop in 5 years then just ask me what I am developing today:

Of course I have learned a lot along the way about not getting too far out over the horizon.  It took an “idea in the shower moment” to get me thinking about the mobile space again and taking the time to do enough research to make sure the idea was doable.  I had also been reading and thinking about the principles of the Lean Startup movement that resonated very strongly with my own good and bad experiences as a startup junky and those two events combined and led me to join yet another startup adventure after properly vetting the idea with potential customers.


Pablo: Which is your experience with mobile development? and with Augmented Reality?


The glue behind bThere’s upcoming products center around mobile location awareness coupled with expert and or time and space critical knowledge. We realized early on that AR had a lot of possibilities to enhance our products in these areas, and the more we thought about AR the more ideas we came up with.

The problem we ran into was the same problem all small companies have, namely how to get access to reasonably priced high quality technology that we did not have the time to build or to maintain, and trust to be enhanced with new features as the market evolved.   

While I used to have a “build it here” mentality  in the current high speed startup environment that is often not possible or even desirable.  The mashable concept is a great force multiplier for startups and one that every entrepreneur should happily embrace. The ability to plug in technologies quickly in order to support a larger idea is critical to success these days. It’s really priceless and lets you do things with a team of a couple of guys fairly quickly that you could not have done even 10 years ago with much larger teams.

We thought we could make use of AR and I was pretty positive about the state of what was available at first since there are a lot of open source, as well as paid libraries and applications available. They all had pretty cool demos that seemed to do just what we needed, at least on the surface.  None of them really worked as advertised though, and the few that came close were either overly complex, way too expensive, or the development staff was completely non-responsive to our concerns. 


Pablo: After this month with the ARBrowser beta running, what is your impression?


We found AR Labs at the last minute and I am pretty happy  that we did.  The location based browser capability took about 2 hours to get a demo doing what we wanted it to do, which in fact itself was pretty remarkable. I was still concerned about how their technology would evolve and how responsive they would be to both any issues we found as well as new ideas that we had, like all developers we needed some tweaks done to get what we really wanted so I e-mailed AR Labs about these issues.

Some of our concerns were simply “didn’t read the docs” syndrome on our part but they were very gracious in explaining those with any “attitude”, while other issues we had were already in the works to be added or fixed, and some of our more future looking ideas were met with genuine open handed interest, some were even put into the development queue. That last part is critically important for small companies that are going to get married to a technology early on because a divorce once you ship a product is pretty messy and never easy.

What the community really needs to start producing great, vs. cute AR products is an inexpensive and dead simple way to make AR accessible to the average developer, the same way Google made map based API’s dead simple and accessible. This common access model will lead to new and interesting uses of AR that will in hindsight seem obvious.

AR Labs are the only ones out there that have done this as far as I can tell and bThere is really looking forward to the product moving out of beta and into production. We also look forward  to AR Labs continuing to push the envelope with new and interesting abilities.  The other offerings they have in the AR space are in areas I had not considered using simply because I had not thought any viable solutions would exist but having used the browser technology  I am fairly excited to try out those offerings as well as this “Smorgasbord” approach suits us perfectly.

Pablo: I’d like to thank so much Todd for his time and the feedback given :)

Do not forget to check out our AR Browser and Image Matching SDKs.

Tags: , ,
Posted on: No Comments

Leave a Reply