Its alot of fun downloading artwork, and pairing it up with machine learning to see how I can stylize my photography. Using Python and a Conda I built an environment on my computer that uses my graphics card to render what the machine learning algorithm sees.
Camera Model
Lumix GH3
Lens
Olympus MSC ED M. 60mm Macro lens
Filter
None
Aperture
f/2.8
Shutter Speed
1/125 sec
Film Speed
1250
Spectrum
Visual Spectrum (VS)
Wavelength
380 through 700 nanometers
Location
North Georgia USA.
The reference photo I used was from a recent photo shoot I did indoors with some plants by a skylight. Using my visual spectrum photo, I ran it through a neutral style transfer application and it outputted different stylized images depending on what artwork I paired it up with.
Using a pattern by an unknown artist It gave the leaves a flame like look. Though in the Alma Thomas art it seems it as alot more trouble stylizing.
The colors transferred quite well using Tullio Carli art. Came out pretty good I think.
The red became dominant just like it is in the Grace Hartigan art. And I think I can see the brush strokes transferred as well.
And as stated earlier the Alma Thomas art I think really confused the machine learning algorithm. It did transfer colors, but seems it painted the whole image in the splotches. If I can get a lua environment working, I can play alot more trained models, bringing new transfer aspects into my photography.
But for right now Python is working well with this app and its been alot of fun. Still finding art that gives the images some really unique looks. I can probably use a couple other trained models but for the most part im limited using neural-style-pt.
Link to software I used: https://github.com/ProGamerGov/neural-style-pt
These images are being hosted by Spee.ch
An image hosting service on top of the LBRY protocol.
Makingblocks.xyz/about
Addresses below to help me buy better camera equipment and support me to travel to locations to do photo and video and overall great blogs in new places. I would be happy to list some of the contributors in my posts for donations that help me along the way.
Coin
Address
BTC:
bc1qhfmvd2gywg4fvrgy2kkkkyqta0g86whkt7j8r7
LTC:
ltc1qdyzm5cwgt8e2373prx67yye6y9ewk0l8jf3ys9
DASH:
XkSqR5DxQL3wy4kNbjqDbgbMYNih3a7ZcM
ETH:
0x045f409dAe14338669730078201888636B047DC3
DOGE:
DSoekC21AKSZHAcV9vqR8yYefrh8XcX92Z
ZEN:
znW9mh62WDSCeBXxnVLCETMx59Ho446HJgq
