images

Zoom In. Now… Enhance! (For Real, Kinda)

The Zoom And Enhance trope has long been the ultimate criminal identification solution and a staple for crime drama television. Its use on screen is often lauded as an example of how Hollywood doesn’t understand technology. The Enhance Button trope simply ignores that the blurry focus and big blocky pixels you get when you zoom in close on an image are the only information that the picture actually contains, and attempting to extract more detail from the image alone is essentially impossible.

Enhance Old Station

Enhance Bank Lobby

However, as a proof of concept, Alex J. Champandard’s Neural Enhance coding project uses deep learning to enhance the details of images. As seen from the gifs above, if the neural networks are well trained, the enhancements are quite effective.

Thanks to deep learning and #NeuralEnhance, it’s now possible to train a neural network to zoom into your images at 2x or even 4x. You’ll get even better results by increasing the number of neurons or training with a dataset similar to your low-resolution image. The catch? The neural network is hallucinating details based on its training from example images. It’s not reconstructing your photo exactly as it would have been if it was HD. That’s only possible in Holywood — but using deep learning as “Creative AI” works and it’s just as cool!

Now let’s vector in and enlarge the z-axis.

via prosthetic knowledge

A Young Globular Cluster

Youthful NGC 362 Globular Cluster

Earlier this week NASA released this dazzling image of NGC 362. It is one of about 150 known globular clusters on the outskirts of our own galaxy, the Milky Way. Globular clusters are giant spheres that contain hundreds of thousands of stars and reside in the outskirts of galaxies. The ESA says NGC 362 is unusual:

By studying the different elements present within individual stars in NGC 362, astronomers discovered that the cluster boasts a surprisingly high metal content, indicating that it is younger than expected. Although most globular clusters are much older than the majority of stars in their host galaxy, NGC 362 bucks the trend, with an age lying between 10 and 11 billion years old. For reference, the age of the Milky Way is estimated to be above 13 billion years.

This image, in which you can view many of NGC 362’s individual stars, was taken by Hubble’s Advanced Camera for Surveys (ACS) If you want a new desktop image, here’s the 42 MB full-size original (it will automatically download).

The Moon 1968–1972

Project Apollo Archive 41

During all six of NASA’s manned lunar landings, astronauts were armed and trained to use modified Hasselblads. During the Apollo missions, NASA’s astronauts took photos of moon landings, moon walks, the lunar surface, the horizon, and the Earth with these cameras. The results included over 20,000 photographs by 13 astronauts over six lunar landing missions. This huge trove of photographs are cataloged at The Project Apollo Archive. NASA also released a large number of these photos on Flickr back in 2015. The photo above is one of my favorites from this collection.

Though shot originally for scientific purposes, many of the photos have an extraordinary aesthetic value that encompasses an inadvertently artful composition. The fine folks at T. Alder Books have sorted through the nearly 15,000 of these photos and came up with 45 images that consist of “unintended artful compositions” and a “beautiful, deft outtake quality,”. The collection will be released in a book entitled The Moon 1968–1972 that will be released later this month.

At a time when archival images are often hastily assembled into digital galleries that get passed around briefly on social media, it’s especially satisfying to sit with an affordable ($18), carefully edited, designed and printed archive of photographs of historical significance and esthetic value. Texts include excerpts from a speech President John F. Kennedy made about the Apollo program, and from an E.B. White story for The New Yorker recalling the first moon landing.

Does A Deep Neural Network Like Your #Selfie?

Best Selfies
Andrej Karpathy trained a Convolutional Neural Network with a dataset of 2 million photographs to determine what makes the perfect selfie. The image above contains the top 100 best selfies (here are the 1,000 best selfies) Andrej concludes that the best selfies have these qualities:

  1. Be female. Women are consistently ranked higher than men. In particular, notice that there is not a single guy in the top 100.
  2. Face should occupy about 1/3 of the image. Notice that the position and pose of the face is quite consistent among the top images. The face always occupies about 1/3 of the image, is slightly tilted, and is positioned in the center and at the top. Which also brings me to:
  3. Cut off your forehead. What’s up with that? It looks like a popular strategy, at least for women.
  4. Show your long hair. Notice the frequent prominence of long strands of hair running down the shoulders.
  5. Oversaturate the face. Notice the frequent occurrence of over-saturated lighting, which often makes the face look much more uniform and faded out. Related to that,
  6. Put a filter on it. Black and White photos seem to do quite well, and most of the top images seem to contain some kind of a filter that fades out the image and decreases the contrast.
  7. Add a border. You will notice a frequent appearance of horizontal/vertical white borders.

Andrej also created a TwitterBot that will judge your selfie. Simply attach your selfie (or a include a link) to a tweet that mentions @deepselfie anywhere in it. The bot will analyze your selfie and give you its opinion (e.g. score 90% means that the Selfie Bot is 90% sure yours would be in top half of selfies. Selfie Bot was not impressed with my selfie.


More here.

My Galaxies

Artifacting My Galaxy

My Galaxies allows you to spell words using images of actual galaxies in our universe. Above you can see “artifacting” spelled out using the respective galaxies listed below (along with their SDSS ID):

A: 587727226230538297
r: 587739721376989338
T: 587738568713109663
i: 587731186736169172
f: 587739132954280095
A: 588011101032546573
c: 587738574072971384
t: 588017979992441297
i: 587739408392323129
N: 587732136987656314
G: 587736979571343696

Sriracha Rooster Drawn Entirely In HTML5 Canvas

The Sriracha Rooster below is rendered entirely in browser. It’s not a gif, jpeg or other image format. I created it using the HTML5 Canvas element, Javascript, and a ton of code.



If you are using Internet Explorer you will have to be updated to version 9 in order to see the image element to work. View the page source code to see exactly how I did it.

Scroll to Top