No, the creepy face above isn’t a still frame of He-Who-Must-Not-Be-Named taken from one of the Harry Potter films. It’s actually a composite portrait of countless faces found in the 2009 James Cameron science fiction film Avatar.
A while back, PetaPixel posted some features about image averaging and faces. Richard Prince created a composite portrait of the 57 faces of girlfriends on Seinfeld. This led to Pat David exploring the averaging of faces with Martin Schoeller’s portraits of celebrities.
I’ve long been interested in image averaging as well; as a measure of central tendency, I like that image averaging can highlight similarities and differences across an array of seemingly equivalent images.
Earlier this month, Noah Kalina released an updated version of his everyday self-portrait time-lapse video, showing how his face changed over more than a decade. The video spans 12 years and 5 months, containing over 4,500 daily photos of Kalina’s face shot from the same angle and perspective.
Than Tibbetts wanted to see what the entire project would look like as a single frame, so he used FFmpeg to extract the frames from the video and ImageMagick to average them. The resulting image, seen above, shows how consistent Kalina was with his self-portraits over the years.
P.S. Last year, designer Tiemen Rapati did the same thing with 500 of clickflashwhirr’s daily self-portraits. That one turned out even better than this one.
Photography enthusiast Sterling Parker created this abstract image by averaging all the photographs he shot in 2007. He tells us,
I have my photos arranged by month, and starting in January 2007, I imported all those photos into GIMP (the freeware image editor) as layers, adjusted the whole canvas to be as big as the largest dimension, then used a custom script to “average” all the layers so each one is an equal relative percentage of the whole. The white background is empty space around photos obviously, and you can see that I took more pictures in landscape than portrait.
After doing that for each month, I averaged all the months together (about 8 months total because I didn’t take pictures some months of that year) and then averaged all the months together. Overall, I’d say this is an average of about 350 photos.
It’s interesting how the image reveals both his preferences for portrait/landscape orientation and also his different camera resolutions.
Image credit: Photograph by Sterling Parker and used with permission
Inspired by Noah Kalina’s viral everyday video a girl who goes by clickflashwhirr has been doing a similar self-portrait-a-day project. Designer Tiemen Rapati decided to make a composite image showing what the average of the self-portraits looks like. Taking 500 images from clickflashwhirr’s Flickr set, Rapati wrote a script that counts the individual RGB values for each pixel, averaging them across the 500 portraits.
What you see here are portraits created by taking photographs of women in 40 different countries and averaging them with Face Research software. It’s not clear how many faces were used for each country, but if you’re thinking that the faces are more beautiful than average, then it might be because attractive faces are generally average. You can also play around with the software yourself, either with existing faces or by uploading your own.