Loading

Crowd Sourced Abraham Lincoln

by AminimalStudio, published

Crowd Sourced Abraham Lincoln by AminimalStudio Jun 19, 2012

Featured Thing!

Description

Crowd Sourcing flickr images to create 3D files.

This is a working proof of concept.

Long story short...
I didn't go to Washington DC to 3D scan the Lincoln Memorial.
Instead, I scanned flickr for all related images to "Lincoln Memorial statue" and downloaded the originals.
By uploading it to 123d catch, I was able to receive a 3D model of Abraham Lincoln's Face.

Recent Comments

view all
Back in 2001, right after 9/11, we actually 3D laser scanned portions of the Lincoln Memorial and created a watertight version of Abe sitting in the chair, which we printed way back then.  

We can't post the data to Thingiverse but if anyone wants to know more, contact us.  See more at: http://www.directdimensions.com/port_featuredprojects.php?fileName=fp_lincolnmemorial
A nifty idea. I wonder if 123D Catch is actually up to the task. I know that on the PC version of 123D Catch you can manually identify "stitching points" to help the program "stitch" the various photo's together. I bet using pictures with different cameras in unique lighting situations will require a lot of manual stitching to help achieve a good result. Did you do any manual stitching to get the face you made?

More from Sculptures

view more

Liked By

view all

Give a Shout Out

If you print this Thing and display it in public proudly give attribution by printing and displaying this tag. Print Thing Tag

Instructions

Here are my plans for a fully automated 3D generator based off of crowd sourced photography.

1. Parse Flickr with Search phrase
2. Download first 100 photos
3. Upload to 123d Catch
4. Download 3D model

Simple enough.
Hopefully google can integrate this into their search engine in a few years (or months).
Back in 2001, right after 9/11, we actually 3D laser scanned portions of the Lincoln Memorial and created a watertight version of Abe sitting in the chair, which we printed way back then.  

We can't post the data to Thingiverse but if anyone wants to know more, contact us.  See more at: http://www.directdimensions.com/port_featuredprojects.php?fileName=fp_lincolnmemorial
A nifty idea. I wonder if 123D Catch is actually up to the task. I know that on the PC version of 123D Catch you can manually identify "stitching points" to help the program "stitch" the various photo's together. I bet using pictures with different cameras in unique lighting situations will require a lot of manual stitching to help achieve a good result. Did you do any manual stitching to get the face you made?
Great idea but the headline photo is hugely misleading about the level of success you seem to have had with this method!
Well, there are problems on my end, and problems on the software's end.

My end:
Some images are photoshopped. I can just add more images.
Also I had more 3D info but only decided to take the face for now.

Software:
It's not made for multiple cameras but managed three or four different ones.
I could align other photos but that would take too long for me.
The software
would need to be tweaked for this type of capturing.
Right now its set up for one camera, one timeframe.
Ha! perfect. this is great. turn flickr into a 3d scanner for monuments and other things that tend not to move. awesome!
Just brilliant!! Anyone up for a few British landmarks? Stonehenge has got to be done :) Would also love to see the Sphinx re-created.
Good work man, love the idea. Instead of google images or Flickr you could try Panoramio. I find the photos of landmarks on panoramio are of a high quality and a variety of angles.

Panoramio might also be simpler from a licensing/copyright perspective as these photos are shared with the intent of documenting a landmark.
What are some other things, places, objects, sculptures that would work?
MLK Memorial.
Winged Victory of Samothrace
Nice. I tried this with Mount Rushmore, but it was hard to find photos from way above...
That was on my list...

But I think you need to print a quadcopter.
Damn, John - this is awesome. Could lead to so many great 3D models available
Great work! :)

And this should work with any stationary object that has a fixed background and sufficient photographed angles. Tourist attractions come to mind: Giza Pyramid, Sphinx, Stonehenge, etc. google.com/search?q=stonehenge
&
amp;tbm=isch
This thing begs the question: who "owns" the resulting 3D model. Unlike #Met3D, these photos were not taken by the model creator, and some photos used to create the model may not even be licensed for use in this manner.

Big ups John, for finding another disruptive dimension to 3D capturing/printing! :)
Not too sure if google images would work just yet.

most Flickr images store the camera information needed to read focus distance information.

Also postion of lights is key. So 24hr changing outdoor daylight is not the best situation.
Luckily, lincoln is semi outdoor and most people take a photo during the day.
I was hoping to get more info on th
e side of his head. Most people photograph that because they believe its john wilkes booth portrait. I just have a giant hole... hmmm...
This will become a fascinating way to reconstruct busts/bodies of historical figures and locations from their photos alone!

Time to scan through archive photos!
Very cool! I love the idea of scraping data to build new models
Top