Tag Archives: image-based modeling

The Era of Post Processing (part deux)

I’ve opened up a can full of thoughts and ideas with my last blog on Post Processing.  So, imagine this:  What if you can create a 3D model out of any image?  What would you do with it?  How would you use it and for what?

This is work I did at MIT with Max Chen and Fredo Durand many years ago.


The Post-Processing Era of Photography

The art in photography used to depend heavily on how and when the photographer captured the light into the film during the photo shoot. I remember counting down the number of shots I had left in my 35mm film, painstakingly developing my black and whites, and using the negatives to create the positives. From acquisition to seeing the image, it took forever. And only after seeing the images (which may be days after the shoot) could you know what you’ve done wrong.

These days, the art in photography depend heavily on the science of post processing. (Can you say Photoshop?) Digital SLRs can pack in tens of mega pixels, at multiple frames per second, at a higher dynamic range (i.e. RAWs), and your instant-gratification desires are gratified. When in doubt, take more photos. And to get that perfect image, you “Photoshop” them or post process using your favorite tool or technique. If you capture enough light, you can (theoretically anyway) recreate and reproject the rays of light from the past to the present.

The quality of photography became directly related to the quantity of photography. I don’t think this is bad at all. The medium has changed — it has become more complex and more expressive, I think (I’m definitely not a “purist”).

So, what type of post processing am I talking about? Take a look at this video above let me know what you think — it was created from a single photograph.


Follow

Get every new post delivered to your Inbox.