This idea is very intriguing. I believe the product is very doable if you have the appropriate tech resources and the right team. I'm unsure of what the target market may be for this application? Middle-aged individuals? And are you planning to make this free because I'm not too sure people will be willing to pay for this unless it has more attractive features. I would love to talk further about this and help you brainstorm some concepts + viable business models. Ping me!

The idea is great !! Its not new though..
Check out http://www.pounce.mobi/ ( it was earlier called BuyCode).

Some positive points
1) Just because it is there, it does not mean you can build it.
2) A business which got acquired and vc backed is a good sign that your idea is already market validated.
3) Because it is validated, you already know what is the market segment you need to target or you can pivot from the known reference point now.. cool ..isn't...

How can you do it ?
1) Ideas are dime a dozen and worth a shit. Don't feel bad. Build an MVP.
2) With the image recognition already democratized using dozens of cloud service providers, you should be able to build something quick with low TCO (total cost of ownership).
3) Identify your business model. You want to copy BuyCode or do something different ?
4) Try selling it - (rubber meets the road). You will feel the heat and smell the burns :-)

Need help ? talk to me.

I love this idea and would use it, don't listen to the naysayers. There have been previous attempts but that should only validate the need.

As a homeowner I would want this, and this would be useful to pretty much anyone looking for non-discount furniture (I say that because the IKEA demographic probably doesn't have time, money, or interest). I hated driving from furniture store to furniture store, and searching online would take many hours.

This would require a huge database of items. Maybe the vendors and manufacturers would give you photos to match against? make your vendors work for you if you can show that you could generate them business.

I think you are going to need some of the following consultants: furniture historian to help categorize older furniture, furniture designer to help sort into categories, a software developer to figure out how to match the image to a database, someone to work with manufacturers, and a lot of people to enter information.

Cool idea. Biggest question is market size. I presume the furniture market is big -- the question will be how many people will shazam furniture and what kind of revenue could you generate from those purchases? Maybe the broader opportunity is to shazam any product you can buy...