My fall adventure

Nate the great · 12-13-2009, 05:41 PM

I doubt anyone noticed, but I was absent a lot this fall. Whole days would go by without a visit to MR. I was gone because I was taking several classes.

I just finished a project for one class, and I'm proud enough that I wanted to share it with you. It's pretty technical, but I bet there are a number of people here who will understand it.

I wrote a program that matched a picture downloaded from Google images with one found in a target photo. For example, I did a Google search on "spam" and then used one of the results. The goal is to find enough similarity between the 2 images that you can reliably say that they show the same thing.

I'm sure you're thinking that this is easy. Did you know that (I think) a quarter of the human brain is devoted to vision? A significant portion of the vision processing in the human brain occurs without conscious thought. Also, it took your brain years to learn how to see objects. This is not an innate ability (I've worked with teenagers, so I know this for a fact). Duplicating this ability is still an interesting problem in computer vision.

some technical detail
The project was written based on the goals set in The Semantic Robot Vision Challenge. The program was written in matlab, and I relied heavily on VLFeat library. I used Harris corners to select the points to try to match, and SIFT to describe the points and match them.

With some refinement, my program is good enough that I'm planning to use it to compete next year. Some success images follow. Enjoy.

dreams · 12-13-2009, 08:12 PM

Quote:

Originally Posted by Nate the great

I doubt anyone noticed, but I was absent a lot this fall.<snip>

I just finished a project for one class, and I'm proud enough that I wanted to share it with you. I bet there are a number of people here who will understand it.

I wrote a program that matched a picture downloaded from Google images with one found in a target photo. For example, I did a Google search on "spam" and then used one of the results.
<snip>.

I did notice as I always try to read your posts. They are not always understandable (like this one) but I figure I'll learn something.

Congratulations on the finished project and Good Luck when you enter the competition.

phenomshel · 12-13-2009, 08:37 PM

This is cool, Nate!! I've often wished there was a scan option to match a picture to google image results...

FlorenceArt · 12-14-2009, 04:39 AM

Sounds like a tough project. I think I read somewhere that Google is working on something like this? It's hard for me to imagine that a computer program can do something like this. Congrats on completing your project!

12-13-2009, 05:41 PM	#1
Nate the great Sir Penguin of Edinburgh Posts: 12,375 Karma: 23555235 Join Date: Apr 2007 Location: DC Metro area Device: Shake a stick plus 1	My fall adventure I doubt anyone noticed, but I was absent a lot this fall. Whole days would go by without a visit to MR. I was gone because I was taking several classes. I just finished a project for one class, and I'm proud enough that I wanted to share it with you. It's pretty technical, but I bet there are a number of people here who will understand it. I wrote a program that matched a picture downloaded from Google images with one found in a target photo. For example, I did a Google search on "spam" and then used one of the results. The goal is to find enough similarity between the 2 images that you can reliably say that they show the same thing. I'm sure you're thinking that this is easy. Did you know that (I think) a quarter of the human brain is devoted to vision? A significant portion of the vision processing in the human brain occurs without conscious thought. Also, it took your brain years to learn how to see objects. This is not an innate ability (I've worked with teenagers, so I know this for a fact). Duplicating this ability is still an interesting problem in computer vision. some technical detail The project was written based on the goals set in The Semantic Robot Vision Challenge. The program was written in matlab, and I relied heavily on VLFeat library. I used Harris corners to select the points to try to match, and SIFT to describe the points and match them. With some refinement, my program is good enough that I'm planning to use it to compete next year. Some success images follow. Enjoy. Attached Thumbnails Last edited by Nate the great; 12-13-2009 at 09:24 PM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
WE ALL FALL DOWN--Thriller	Simon Wood	Self-Promotions by Authors and Publishers	5	06-09-2011 03:42 PM
The ADVENTURE OF ANNA THE GREAT a swashbuckling adventure for girls	DaringNovelist	Self-Promotions by Authors and Publishers	1	10-11-2010 11:25 PM
It's already Fall	Falcao	Amazon Kindle	8	09-28-2010 01:23 PM
Unutterably Silly The fall of the UK	SameOldStory	Lounge	55	09-15-2010 08:14 AM
Fall for the Book	Nate the great	Lounge	11	09-24-2009 02:44 PM

12-13-2009, 08:37 PM	#3
phenomshel ZCD BombShel Posts: 4,793 Karma: 8293322 Join Date: Jan 2009 Location: The Frozen North (aka Illinois, USA) Device: iPad, STB Kindle Oasis	This is cool, Nate!! I've often wished there was a scan option to match a picture to google image results...

12-14-2009, 04:39 AM	#4
FlorenceArt High Priestess Posts: 5,761 Karma: 5042529 Join Date: Jul 2008 Location: Montreuil sous bois, France Device: iPad Pro 9.7, iPhone 6 Plus	Sounds like a tough project. I think I read somewhere that Google is working on something like this? It's hard for me to imagine that a computer program can do something like this. Congrats on completing your project!

Advert