Patent • Olinda: Create Location-Specific Artwork With AR


This was a side project inspired by the street art/murals seen in Mission district in San Fransisco. I returned from there thinking there must be a way for artists all over the world to collaborate digitally without actually being physically present.

Later this became my Adobe Kickbox project.


1. Concept

I came up with a workflow for creating, modifying and viewing location-specific artwork using augmented reality (AR). Here, AR would be used to project layers of digital information onto the physical environment. A mobile app called Olinda was designed to make it consumer-centric.

There were 3 steps:

  1. Create a canvas from the built environment
  2. Create the digital artwork on the canvas, and post it to the physical location
  3. View the artwork through an AR-enabled device

Screens from the workflow

olinda wfOlinda app as a hub for social street art creators and consumers

2. Communication

When a concept doesn’t solve an existing problem, but envisions a future value, it’s challenging to tell that story in a simple way. I tried a number of different ways:

a) Olinda film

The goal was to show that the app could be used by anyone for making an experience really special. I scripted the story and the screenplay, gathered actors and scouted for locations and equipment, shot the film, directed the music and edited the film. With a little help from my friends!

This slideshow requires JavaScript.

b) Digital presence

I worked on two types of websites, presenting the concept to two types of users: the creative pro and a typical consumer|

c) Social Media Campaign

I also launched a Facebook Ad Campaign and gathered a few likes:



Overall it created quite a buzz internally, but the actual technology was pretty complex to build in 2013. However I was able to file a patent (AD01.3322US01).

Date of Project: May 2012-June 2014

My Role: Everything



Patent • Derrida: Deconstructed Video Navigation


Another side project that was born from the challenges of viewing video file without knowing much about its content beforehand. Our aim was to create either

  • a new navigation system, over the sequential timeline
  • a new video ad unit for Adobe Primetime that would build better brand engagement.

At the time I was quite inspired by the work of Ellsworth Kelly and Karel Martens.


We explored video navigation using existing short films from Youtube and came up with

(1) An algorithm to generate the content visualization of a video

(2) A number of variations of these visualizations, to enable alternate non-linear navigation apart from the timeline.

We worked on about 50 videos of duration from 30 sec to 1.5 min. We found that any of these videos could be deconstructed to components which are then visualized in different ways to give different viewing experiences.


We took this video as an example to show how many this works.

This video can be deconstructed into its components and then recombined to tell variations of the same story to different types of users.

Derrida 00

Maybe a repeat user might see a different version the second time. This would enable digital marketers to repurpose existing content to build further engagement.Derrida 01Derrida 02

Derrida 03

Links to other engagement could also be embedded. For example from this navigation, users can directly go to buy the displayed merchandise, download additional media, or learn more about the location, etc.

Derrida 04


Patent: P5209US01: Deconstructed Video Ad Units

Date of Project: Jan 2014-June 2014

My Role: Designer & Product Owner

Team: Nutan Sawant and K Balaji