photoriot

Alignment makes random pics work together more-predictably.

I want my pics to generally have any vertical line in the center of the frame be aligned vertical. What I've been doing is manually rotating and cropping, but I've got too much backlog to catch up on. At the same time, I'm sold on the concept of mirrorless now, and it seems like a mirrorless cam should be able to rotate and crop live in the viewfinder and when saving the photo.Failing that, I'd want the tilt recorded in metadata so I can automate the rotate/crop on the computer. Does the RP or other R's record tilt? Ideally, I'd like to see a maximal crop rotated in the viewfinder. I'm told the 5D's record tilt, so fallback is to get a beater to take a bunch of reference pics to train neural nets to detect the angle.

http://phobrain.com/pr/home/gallery/pair_horiz_blinds_shadows_ofc_bench_armrest.jpg

Street photography.

Done! :-)

The url got hidden, demos.algorithmia.com - probably it could be much improved with effort, and I assume there is much better proprietary stuff used for colorizing moves.

Have you power-cycled everything?

Google has wrapped their deep learning image recognizer into a package you can play with fairly easily if you can handle console access to a computer ("TensorFlow for Poets"). You simply create a folder for each thing you want to recognize, and fill it with examples (~600 pics per flower in the flower classification example they provide), and run a command for 5 mins to a day (2012 laptop). Then you run another command to get the classification for a new image. The deep parts of the net remain fixed, and were trained by Google on a big database of pics - it just fine tunes the final stage to your pics. Here are my first results - pretty shaky for my own purposes, but not bad for a start. I have different requirements than Google, e.g. if I say the photo has red or white, I mean it jumps out at you, not simply that it is there, so to get the most out of this technology I might need to get a GPU or 4 and retrain the entire network for a week or two, and manually classify way more than the 15K images I have now as input to the algorithm.

Note how it gets Weston's cabbage red. Colorize Black and White Photos http://edward-weston.com/wordpress/wp-content/uploads/2015/01/Cabbage-Leaf-1931-39V.jpg https://a.1stdibscdn.com/archivesE/upload/a_115/1485891672037/johnny_edward_weston_1944_weston_gallery_carmel_california_l.jpg http://edward-weston.com/wordpress/wp-content/uploads/2015/01/Pepper-1930-30P.jpg http://edward-weston.com/wordpress/wp-content/uploads/2015/01/Nahui-Olin-1923-16PO.jpg

I haven't read the article in full, but suspect the author could benefit from reading this: Photo Editing with Generative Adversarial Networks (Part 1) | Parallel Forall

Another concept that the photos don't support, again planned meticulously long in advance of just loading the pics in my manual pair formatter.

Can you believe it took me 3 days to add a 'Let pics repeat' option to Phobrain? This demo pair seems like something of a match for the pics above.

..Bueller? Absent any failures, I wonder what bulges the plastic out like that. We may wish our model to predict a different amount of variance in y for different values of x. This is called a heteroscedastic model. -- The Deep Learning Book

I have had this pair in mind for maybe two years, and finally got tired of waiting for Phobrain to generate it for me automatically, and it is a case of be careful what you wish for. Other failures long in the making?

I thought the nurse was saying it about the ward's pictures. "A nurse saw my photos and said, they are like the photos on the ward - they are all of nothing!" does someone calling photos "fine art photos" necessarily mean they think they're good? True, they can be fine art and not good. Certainly mine weren't stellar enough for Karp to invite me to give a show - he said he was booked for 3 years or something. But it seemed like he saw what I was shooting for. I agree that hobbyists can easily be more interesting than artists, and high school plays and musicals are usually more fun than professional versions for me (small sample size), so I'm going for interest rather than quality, those moments I feel like I'm seeing something completely new.

I like how my mind puts together the correspondence between the shadow plant and the real one, as if they are magnets and my neurons are iron filings following the field.

A different or further diptych turns it again. Hello, Bill. You might like Phobrain even better Julie, if your head doesn't spin off :-) Latest head-turning technology has to do with the sequencing of the now ~20K curated pairs. When you click on the left pic, it associates at least one keyword between one of the current pics and one of the next pair, giving the chance for themes to emerge, sometimes giving the impression of statements or commentary. When you click on the right-hand pic, it mixes in strong, 4-way keyword matches with color-based matches among the pairs that are most weakly connected by keywords, using view time and mouse behavior to select which approach to take, which interactivity I liken to throwing a stone into a pond.

Or you could be missing what she's seeing, which might just be bad pictures. Since she said it about my pics (pre-diptych), you can be the judge too. I compare it to hearing Ivan Karp of OK Harris Gallery in NY, say "This is fine art photography" after going through a couple of my books, and cling to my illusions. :-)

So, 'nothing' is both a quicksilver category spilling into your irises and out of your fingertips, and a value judgement within some context that may be misjudged, i.e. it's easy to project on nothing.

These bits of pavement were diligently located or recreated, and placed in the Smithsonian because they were mentioned on photo.net, an early instance of a pre-quantum internet site, pursuant to Trump XXXiV's order that everything must be something.