Panacea

Post on 04-Jul-2015

811 views 2 download

description

it's an Augmented reality android app which can be used inside an enterprise to manage contacts and location services

transcript

Augmented Reality?

Overlaying of digital data on the real world

Navneet

The

Team

http://tinyurl.com/twdemo

An augmented reality application to be used inside an Enterprise to manage contacts and location

services.

We built

“ “

PANACEA

TWANAC

EA* Possible names for the application

Contacts Location

Contacts Location

Contacts Problems

Not Updated

Unmanageable

Non-scalable

Contacts Problems

Exposed Phone numbers

What happens if the Phone number changes?

Wait! Is there a name of a person as well? So, now I know

the person as well his/her phone number.

Contacts Problems

Navneet KumarDeveloper

ThoughtWorks Technologies (India) Pvt Ltd2nd Floor, Tower C, Corporate Block, Diamond District Airport Road, Bangalore - 560 008, Indiatel : +91 80 4064 9570, fax : +91 9686577076navneetk@thoughtworks.com

… passé

Contacts Problems

Navneet Kumar

ThoughtWorks Technologies (India) Pvt Ltd2nd Floor, Tower C, Corporate Block, Diamond District Airport Road, Bangalore - 560 008, Indiatel : +91 80 4064 9570, fax : +91 9686577076navneetk@thoughtworks.com

Future of Visiting Cards

. . .You know why?

Contacts Problems

My only use for it

now is to attach it to

gifts that I give.

vCards 2.0 A radical new way of exchanging

contacts inside the enterprise

Contacts Management

You meet these guys at Dev-Camp. How long do you

think it will take you exchange your contact

information?

We say about 10 seconds!

* Stock photo from stck.xchng

PostersA new way of putting up Assistance

and event posters.

A new way of putting up assistance posters. No numbers, no hassles.

Event Posters - Revisited

* Photo courtesy thoughtworker.com

Presence

@starkcoffee - hopefully my.thoughtworks will be a completely fresh experience to TWers !

* VISION

Event / Conference Feedback(1/2)

* VISION

Event / Conference Feedback(2/2)* VISION

Comments Feedback

Talk: Introducing my.thoughtworks!

TechHow we do this.

Application Walkthrough

- 3 step process

- Core of the idea is a ML

Algorithm for face recognition

- As with any ML algorithm, the larger the training set, the better the quality of the algorithm.

Face Recognition

Status flags convey what thread is currently active and

processing.

Step 1 Detecting…

No Faces in this Image

Step 2 Scanning…

A Face Found !!!

Step 3 Fetching…

Fetch Data From Active Directory

Pratekhsa UdayPrateeksha@thoughtworks.comMobile : 9686577076

Uh Oh!Implementation and Challenges

First Approach

Face Detection

Training Set

Face Recognition

App

Massive Database of Photos

LDAP

Problems with this approach

Face Detection

Training Set

Face Recognition

App

Too Painful

Not Real time – Not scalable

Second Approach

C++ Libraries

Compile Using

Android NDK

Gives

opencv.so

opencv.so

Java Native Interface (JNI)

App

Face Recognition

Pros & Cons - Second Approach

opencv.so

Java Native Interface (JNI)

App

Near Real time

No Service Calls

Totally in Phone

Not Accurate

Poor Training data

Face Recognition

Local Remote

Face Detection

Hybrid Approach

Face Recognition

Training Mode

Scanning Mode Face?

Yes

Train Every Employee ?? Damn ! It’s painful

Distributed Dataset Training

Contacts Location

Do you know these

applications?

* Logos courtesy their corresponding websites

Location Problems

Only 13% of the smart phones sold across the world are Location Aware

Less than 4% of all mobile phones sold last year were Location aware.

* Stats courtesy: http://textopiablog.wordpress.com/2010/02/22/how-many-people-own-a-location-aware-device/

Location Problems

Too many places. Search doesn’t solve the problem

either. *screen shot courtesy google.com

Location Problems

Chinese Restaurant

Pizzeria

Cafe

Roof top - Bar

Roof top Restaurant

Multiple Avenues – Same GPS Location

LocationCreating and using GPS agnostic

Hyper local communities

Introducing the PlaceMark!

Hyperlocal Community

ChicagoBangalore

Chennai

Pune

Melbourne

Enterprise Network

Face Detection

Implementation

Face Recognition

Scanning Mode Face?

Yes

Location?

Main UI SurfaceOverlays

Gab Location

Frames Detection Recognition Services

Info Overlaystatus Overlay

Application Architecture

Cam

era

Android native

detection

Recognition service Gab

serverLocation

server

Pro

cess

ing