Visiting India has always brought lots of joy as I get to see how my homeland is fairing over the time I haven’t been here (haha, I talk like a condescending NRI already ;) ). Also, it is this vacation time where I try to meet my relatives and as many friends as time permits over the short period. One of the special things that happened over my current visit was the chance encounter with an old professor of Electrical Engineering from my undergrad university IIT BHU. He retired after a career that spanned over 4 decades and was one of the most respected figures among our faculty members at BHU.

Move to New York City was the highlight of the past year for me. It had been exhilarating not only for the fact that I moved to the Big Apple but also that it was my first job ever. Like ever means I have never been to any industry/internships ever. I anticipated this big change to affect me in a multitude of ways … and it did!

In my understanding of the world, I find myself dwelling a lot these days on the philosophy of Vedanta. The very nature of this philosophy is different to what we are used to hear and think.

After graduating from my Masters university, I got a chance to get involved with the local Toastmasters group in Minneapolis. Here is the ice-breaker speech that I delivered in front of the members.

I am trying a very ambitious problem in one of my Robotics Course project. The aim is to land a UAV on a moving UGV using Reinforcement Learning. The idea behind the whole process was quite vague so I started using my course project as a way to familiarize myself with the ongoing RL research.

In the previous 2 weeks or so, I was working on some aerial imagery of farms. One important aspect that stands out in these images are, when you view them from such great heights, they appears as big images with repeated miniscule patterns very similar to textures. In my research, we intended to find the orientation of textures generated this way. I tried digging up as much as I can in various forums for finding the texture orientations. While there are numerous research papers on generating a orientation map but sadly it is difficult to reproduce that in code.

I got interested in Homography a few days back since it was needed for my research. So I though why not do a simple tutorial showing how to use OpenCV to generate a basic panorama.

Hey Guys. I have spent days if not months installing CUDA 7.0 and CUDA 7.5 in Ubuntu 14.04 and Ubuntu 16.04 over different laptops(Dell and Asus). I finally got them working. While there are numerous tutorials present over the web, most are incomplete and error prone. The difficulty comes in installing the Nvidia drivers. If one does a mistake then the system crashes(since you need Nvidia drivers for Cuda only but not for graphics display. The default settings messes the OpenGL display drivers) and you need to do a fresh install (again there are many messy ways to escape without a fresh install but lets try to make it correct from the start anyways).

I had been trying to figure out a particular idea, that I think many of us face at 22 yrs of age as graduate students in USA - Should I go for a PhD?

It had been a long time since I came here. Our Summer vacations have started and I am trying to kick start my todo list for summers (which is another post just so that I can remind it to myself). I got a roadbike and am lately enjoying the biking experience along the beautiful trails present in Minneapolis.

This happens to be the one thing that puts me into conflicts a lot these days. A conflict that is within me and is against myself.

These days I have been consumed by the notion of making myself antifragile. This particular term arises from one of the Taleb’s book.

Being a Masters student in the States, if I were to write about any of the topics as my first blog post … this should be the first one.

Time flies isn’t it. It doesn’t wait at all and all you are left with is the memories which are more like a blur.

Understanding the concept of Libraries is important if you have to deal with C or C++ code on a daily basis. For me, since these days I am working with OpenVX specification and taking the Operating Systems class, it has become an absolute necessity.

I will try to get you set up with OpenVX sample implementation in your system. The following goes on for Ubuntu(14.04). (Though it should work for any unix based system.)

I have successfully implemented the triplet clustering method from the 2nd paper. I am able to follow Abinash's comments this time. He also appears happy with my coding so far. The results are all cool :)

I have moved on from Wordpress to here as I found myself very cluttered in that place. Github appears more like a cozy and homely place to reside! :)

Woah! I am enjoying every bit of coding now. I have figured out the whole algorithm from the basics! I had been a fool all the time :) . Though I still think that the current understanding couldn't have been possible if I didn't made all the mistakes! So Thumbs Up!

The Mid-term is over. I am through the first half of the Google summer of Code! As far as the accomplishments are concerned, I have almost implemented the culmination of 3 papers in python and have kept on updating my pull request here: pull 420 .

Continuing from my previous post on being able to understand the gist of the culmination of the 3 papers, I think I have some ideas now. Not 100% still. Being reading the papers and so much of mathematics related to it, I thought giving in to write code side by side will be good.

Hi all!

The community bonding period is over. I tried to start with understanding those source codes which have implemented the same class of algorithms that I am hoping to implement over the GSoC timeline.

Today I will be discussing some of the ways in which inference algorithms especially the Linear Programming ones work in OpenGM.
So there are many ways to solve an energy minimization problem.
From the table below we can get an idea about the different accumulative operations used to do inference:

I had been trying hard to get accustomed with the OpenGM library so that I get to understand what are the things that I will be needing for my implementation of Approximate algorithms.

It had been quite an awesome thing to be selected in Google Summer of Code for the second time. Along with it, I also got an offer to pursue masters in Johns Hopkins University in Computer Science in upcoming Fall. The god has been great to me! I thank almighty for all this .

At some point of time in my undergraduate studies, I got bored of the structure of life we go through. I think it was related to less mental simulation that we get used to while we are there. However, these less mental simulations can be fruitfully converted into some beautiful gems as captured by me and Kundan in the following photo-story.

I have made some codes for converting Mat objects from OpenCV library to CDenseFeatures object in Shogun Library. There are basically three methods of conversion.

My work is primarily to develop a clean way to convert between OpenCV’s Mat object into our Shogun’s SGMatrix object. This would be really cool, once we have it as there will be a seamless interaction between both the libraries. Whenever we need machine learning capabilities, we can switch to Shogun and whenever we need the Image processing manipulations we can work on with OpenCV.

This week turned out to be pretty amazing! Had been working and enjoying a lot with my mentor. He is a super-cool guy. Hope to meet him one day!

The last month proved quite terrific for me. I got selected in the Google Summer of Code 2014 for Shogun Machine Learning Toolbox. I hope to have a fun time this summers with lots of Computer Vision stuff and coding. It's always good to flaunt with shiny things authored under your name :).

For last 3 months,  me and my two other friends  worked hard for this event. We needed to provide our juniors a platform from where they can learn. The gluttony of the events held presently in Prastuti always made me sad. To get over all these things, we brought Perceptron.

It happened when I was in the 2nd year of my college at IIT(BHU), Varanasi. We prepared to participate in a technical fest of IIT Delhi named TRYST. The compedition was for a maze solving robot. There were 7 of us in the team. We were totally different people having separate skills and experiences in bot making. We were all excited to be together there to feel the extravaganza. So, this is how it happened…
We started working from 1 month before. Couldn’t completely benchmark our bot against the conditions that we were meant to face at the competition. We spent 3 days in Delhi, mostly debugging.