More adventures with Photoshop’s new “Super Resolution” machine learning algorithm, which quadruples the resolution of a processed image: a view of Berkeley and San Francisco from Grizzly Peak.
Extracting additional information from an image by “enhancing” it has long been a ridiculous trope of police procedurals; it’s with great amusement that I noticed that Photoshop’s new “Super Resolution” capability (which uses machine learning to quadruple the resolution of an image) is under an option called “Enhance”. The first subject for enhancement was this picture I took of Berkeley and San Francisco in 2011. It’s worth the click-through for the full resolution version.
(Adding to the super-resolution theme, this image also contains, in the lower-center, the Molecular Foundry and its associated center for electron microscopy.)
We’re not far from the ninth anniversary of the founding of Decaseconds, and I came upon this arresting image of Berkeley Marina, the Marin Headlands, and the Golden Gate Bridge (never before published here) while searching for just the right anniversary shot. I’m fascinated by the way the orange emission of sodium vapor lamps lighting Berkeley (I’m sure now all swapped for white LEDs) matches the Rayleigh-scattered oranges of the winter sunset. The same wavelengths of light, coming from completely different mechanisms.
The year 2020 is here! “Cautious optimism” remains my default lens for the future, but a look back over the photography of the past decade (like this shot from the Molecular Foundry overlooking San Francisco during my sabbatical), I’m feeling a bit excited. The first major upgrade in my shooting platform is planned for 2020 (the Nikon D7000 is getting a well-deserved retirement) and I can’t imagine the improvement I’ll see when I jump an entire decade forwards in camera technology.
My favorite view of the Bay Area (and the view that first let me define the idea of the civilization gradient as an element of my photography) is layered up with loads of detail. Down in Berkeley Lab is the building where I worked on sabbatical, and across the Bay Bridge is the completed Salesforce Tower hiding in the marine layer. The differences, particularly from the last time I showed a very similar shot from the spring, are in nature: the high-altitude clouds have been replaced with empty skies and that rolling marine layer, while the green hills have shifted to a dry, highly flammable tan.
I sometimes reprocess older pictures when I find some new approach or something special in an image that wasn’t there before. This picture is a bit different—though I captured it at this time seven years ago, I found that I felt no urge to reprocess any part of it. I was happy with it then—though apparently not enough to post it until now—and I’m impressed with it today.
Two of my past St. Lawrence University students are working on their Ph.D.s at Berkeley and I discovered yesterday that one was giving her Graduate Research Conference (Berkeley’s version of a thesis defense, but earlier) while the other was in the audience. I’m very proud of both of them.
Understandably, this had me thinking about my experiences at Berkeley. In this picture from Grizzly Peak, the perspective folds together Oakland, San Francisco, and Berkeley. In the foreground, look at those gnarled trees—they’re weird but they’ve grown tall. I’ll take that visual metaphor for the grad school experience. I took this picture on Christmas Day in 2016, so I guess that makes these Christmas trees, too.
I took this picture two years ago, during a wonderful springtime in Berkeley when a rainy winter had made the hills lush and green. The view is enormous, overwhelming: Oakland, San Francisco, Emeryville, and Berkeley all packed into one. I like the contrast of the tiny path on the green hilltop on the left side of the image providing a quiet contrast.