Artist Daniel Ochoa, 2018
This project grabs images from google street view api and classifies them into buckets one, on left(or top) and zero on right(or bottom).
The significance of this technology is an algorithm can sort hundreds or thousands of images and bucket them for their
specific need. In this application
of the build, 10 images from each bucket are randomly placed next to each other.
These images were classified by an algorithm not a person.
I labeled a training set of images about 150 for each bucket as one, or zero.
Images placed in bucket one generally had high contrast and a facade of a house was dominate.
Images placed in bucket zero are low contrast or obscured by rain on lens. The image is dominated by the street, cars,
brush or other imagery that was not a facade of a structure.
Transfer learning was used with the Inception
v3 Convolutional Neural Network
with tensorflow to create a model locally.
The genearted model can classify images not in the training set as one or zero.
754 images were classfied and the data is stored in a JSON file. Every 10 days images are pulled to the page via the google street view api.
I am using a version of this tool to select images from Oakland, CA to create
a body of paintings. Oakland Street View
New machine learning technology toolkits like tensorflow combined with vast amounts of data(images from google street view)
gives rise to complex reality thresholds.
Essentially the images can be transformed into meta data that does not have to be seen visually to be categorized. The
preferences of the user is captured in the algorithm.
Through history a key factor in art movements and aesthetic changes are caused by technological changes. For example, commodifying paint, more
colors with reduced prices,
and putting it into tubes gave rise to impressionism in the late 19th Century. This put artists outside and in direct
contact with the environment they were documenting instead of inside
their studio to spend time with subtle color changes and explore the physics of color through transferring natural light into paint. Which
in itself is an abstraction because color mixing of light and color mixing on the pallet are not equivalent.
The commodification of data and machine learning tools is similar to the flash point of what oil paint in tubes did for impressionism. New tools, new results.