[Jderobot] GSOC'15- Project Idea 7,8,9

[Aman Verma]

Dear Sir,
My name is Aman Verma and I am final year undergraduate student from
CHARUSAT university, India.

I have got interest in Machine learning, Artificial intelligence and
robotics and had been building robots from my very first year of under
graduation on platforms like Arduino and Raspberry Pi.
My works also include Image processing based on Opensource language
"Processing" and have been an active developer since. Moreover I have used
OpenCV library as well. I have developed few android apps as well and
created various augmented reality desktop applications based on marker
detection in real scenes.My projects include Surface computer, Gesture
recognition using capacitive sensing, Object tracking in real time scenes,
face tracking ,feature extraction etc

I am being right now an Intern at TATA consultancy Services(TCS) with just
one month left and designing a middleware Framework.I had previously been
intern at a local startup where i built a home automation product using
gesture Recognition on Kinect. You can have look at a video that i posted
on YouTube that shows various gestures that I implemented using MS Kinect.
Gesture Recognition Using Microsoft Kinect
<https://www.youtube.com/watch?v=zj0AWx6Vsb4>
I have done courses on Machine learning and Artificial intelligence from
coursera.org
I will showcase my projects soon.

After following the ideas list on your page,I feel interested in project
idea

*#7. Project: Development of Kinect2 programming tools*
*#8. Project: Structure reconstruction using kinect2 device**#9. Project:
Object recognition using kinect2 device*
for GSOC15 which match superbly with my skills.
However, I have got few questions in my mind which i think i should
consult with you,
#1 For the project idea 7, since we can get the RGBD sensor data from the
kinect , the problem that we are going to solve is to properly align the
color information and depth information of the points available before
the camera,since the actual sensors
are separated by some physical distance, it gets difficult to get all the
information of the image returned by the both sensors of specific points.
Am i rightly interpreting the project idea ..?

#2 For the project idea8, since Kinect have a limitation of
detection up-to a distance of 5 meters , how the map could be created
for objects or area which fall out of its range? Also, will the Kinect be
kept stationary at a place while mapping is done or it could be moved with
suitable algorithm to store point cloud based on its movement.?
#3 I know that kinect2 is vast improvement over the original Kinect
from xbox360, is it compulsorily to use kinect2 with these projects.?

Awaiting for a kind response from your side.

Regards
Aman Verma
------------ próxima parte ------------
Se ha borrado un adjunto en formato HTML...
URL: http://gsyc.escet.urjc.es/pipermail/jde-developers/attachments/20150316/67ebe3fb/attachment.htm