Grokstyle wins LDV Vision Summit 2016 Entrepreneurial Computer Vision Challenge

Entrepreneurial Computer Vision Challenge Winner: Grokstyle, Sean Bell, CEO & Co-Founder ©Robert Wright/LDV Vision Summit

Our annual LDV Vision Summit has two competitions. Finalists receive a chance to present their wisdom in front of hundreds of top industry executives, venture capitalists, top industry executives and companies recruiting. Winning competitor also wins $5,000 Amazon AWS credits.

1. Startup competition for promising visual technology companies with less than $1.5M in funding?

2. Entrepreneurial Computer Vision Challenge (ECVC) for any Computer Vision and Machine Learning students, professors, experts or enthusiasts working on a unique solution to empower businesses and humanity.

Competitions are open to anyone working in our visual technology sector such as: empowering photography, videography, medical imaging, analytics, robotics, satellite imaging, computer vision, machine learning, artificial intelligence, augmented reality, virtual reality, autonomous cars, media and entertainment, gesture recognition, search, advertising, cameras, e-commerce, visual sensors, sentiment analysis, and much more.

The Entrepreneurial Computer Vision Challenge provides contestants the opportunity to showcase the technology piece of a potential startup company without requiring a full business plan. It provides a unique opportunity for students, engineers, researchers, professors and/or hackers to test the waters of entrepreneurism in front of a panel of judges including top industry venture capitalists, entrepreneurs, journalists, media executives and companies recruiting.

In the 2014 and 2015 Summits the ECVC was organized into predefined challenge areas (e.g., "estimate the price of a home or property," "estimate how often a photo will be re-shared") plus a "wildcard" category.

Initially we proceeded in the same way for the 2016 ECVC, but we found that the most exciting entries were overwhelmingly the wildcards, so we decided to go all-in on that category. Attendees at this year's summit bore witness to the outstanding lineup of finalists, including GrokStyle (visual understanding for interior design) from Cornell, MD.ai (intelligent radiology diagnostics) from Weill Cornell, DeepALE (semantic image segmentation) from Oxford University and Vision+Love (automated kinship prediction) from Carnegie Mellon.

Congratulations to our 2016 LDV Vision Summit Entrepreneurial Computer Vision Challenge Winner: Grokstyle, Sean Bell, CEO & Co-Founder  

©Robert Wright/LDV Vision Summit

What is GrokStyle?

GrokStyle, co-founded by Cornell researchers Sean Bell and Kavita Bala, is developing state-of-the-art visual search.  Given any photo, we want to tell you what products are in it, and where you can buy them.  We want to help customers and retailers connect with designers, by searching for how others have used and combined furniture and decor products.  The world is full of beautiful design -- we want to help you find it.

As a Ph.D. Candidate - what were your goals for attending our LDV Vision Summit? Did you attain them?

My goals were to understand the startup space for computer vision, to connect with potential collaborators, find companies interested in building on our technology, and generally get our name out there so we can have a running start.  The event definitely far exceeded our expectations and we attained all of our goals.

Why did you apply to our LDV Vision Summit ECVC competition? Did it meet or beat your expectations and why?

Serge Belongie recommended that we apply, and saw the value that the summit would have for us.  We were excited, but certainly did not expect the amount of positive feedback, support, and connections that we made.  My pocket is overflowing with business cards, and I’m excited to continue these conversations as we turn our technology into a company.

Why should other computer vision, machine learning, and artificial intelligence researchers attend next year?

I think that all CV/ML/AI researchers should attend events like the LDV Vision Summit.  The talks here are interesting and varied, and it is inspiring to see how algorithms and computer vision research are having a real impact in the world. You don’t get that at academic conferences like CVPR.

We try to have an exciting cross section of judges from computer vision experts, entrepreneurs, investors and journalists. Asking a question is Barin Nahvi Rovzar, Hearst, Exec. Dir., R&D & Strategy. Judges included: Serge Belongie (Prof., Cornell Tech, Computer Vision), Howard Morgan (First Round, Partner & Co-Founder), Gaile Gordon (Enlighted, Sr. Director, Technology), Jan Erik Solem (Mapillary, CEO), Larry Zitnick (Facebook, AI Research, Research Lead), Ramesh Jain (U. California, Irvine, Prof., Co-Founder Krumbs), Evan Nisselson - LDV Capital, Partner),  Nikhil Rasiwasia (Principal Scientist, Snapdeal), Beth Ferreira (WME Venture Partners, Managing Partner), Stacey Svetlichnaya (Flickr, Software Engineer, Vision & Machine Learning), Adriana Kovashka (U. of Pittsburgh, Assist. Professor Dept. Computer Science) ©Robert Wright/LDV Vision Summit

What was the most valuable part of your LDV Vision Summit experience aside from winning the competition?

The most valuable part of the summit was connecting with three different companies potentially interested in building on our technology, and with four different potential investors/advisors.  Last year, a key potential collaborator had presented at LDV Vision Summit, looking for computer vision researchers to solve challenging problems in visual search, interior design, and recognition.  This year we were able to connect and say “we solved it!”

Sean Bell, CEO & Co-Founder of Grokstyle ©Robert Wright/LDV Vision Summit

Do you have any advice for other researchers & PhD candidates that are thinking about evolving their research into a startup business?

My advice would be to keep potential commercial applications in mind, early on in the project, so that what you end up with at the end is easier to take out of the lab and sell to the world.  For me, one of the most challenging aspects of research is deciding which problems are solvable and which are worth solving -- if you are interested in startups, this is even more important.  There is the extra step of understanding who cares and who wants to use it.

What was the timeline for you to take your idea for your research to evolving it into a startup plan?

We presented a research paper at SIGGRAPH 2015 about our ideas from last year.  It has taken us a year to flesh out the work, develop it from a research prototype to a product prototype.  But there is still a lot to do.  I am graduating in a few months, and Prof. Kavita Bala is joining full time on sabbatical.  We plan to hit the ground running this summer with our engineer Kathleen Tuite, and two interns we are taking on.  As technologists, we are looking to partner with business people to take the lead on evaluating which markets and customers can benefit the most from our technology.  Starting in the fall, we plan on fundraising to help scale up our technical infrastructure.

Judges for our Entrepreneurial Computer Vision Challenges ©Robert Wright/LDV Vision Summit