This hands-on workshop will begin with introducing traditional computer vision techniques based on visual pixel processing, with a focus on tracking bodies, faces and hands. The workshop will build up to modern computer vision based on machine learning, allowing for complex analysis, such as body pose estimation. From there, we'll dive deeper into machine learning as a broad field, covering related techniques for working with other media such as text and sound.
Dates: Two-day workshop meets 6pm-9pm on Monday, May 13th & Wednesday, May 15th, 2019
Cost: $240
Experience Level: This workshop would be ideal for participants who are moderately skilled at programming and want to learn how to use machine learning to manipulate visuals, text, and sound.
Requirements:
Basic understanding of JavaScript.
Bring a laptop.
Will cover a range of topics including:
• An intro to machine learning concepts including supervised vs. unsupervised learning, regression vs. classification, clustering, and dimensionality reduction.
• How images are processed: everything from pixel manipulation to high level analysis by neural nets.
• Image sources, addressing variations in different cameras.
• Some recurring cultural questions raised by computer vision, machine learning and the rise of AI-assisted automation.
“Computer Vision” refers to a broad collection of techniques that allow computers to make intelligent assertions about what's going on in digital images and video. Thanks to recent advances in affordable vision technologies (such as Kinect, IR webcams, and Leap Motion), and armed with a slew of powerful but simple heuristics (tricks!) for extracting useful information from images -- a large number of artists have begun to explore the new possibilities for interaction made possible by cameras.
“Machine learning” refers to explaining tasks via examples (training data) instead of instructions (code). Machine learning has come of age in the last six years. Fueled by the rise of new algorithms, new hardware, new toolkits for efficiently solving complex problems, and huge datasets compiled from everything from sensor networks and surveillance cameras to social media. In 2012 we saw the first algorithms able to effectively identify objects in everyday images from a large number of possibilities, and within three years the research had moved on to generating complete captions for those same images, or even generating low-resolution images from the captions.
Kyle McDonald is an artist working with code. He is a contributor to open source arts-engineering toolkits like openFrameworks, and builds tools that allow artists to use new algorithms in creative ways. He has a habit of sharing ideas and projects in public before they're completed. He creatively subverts networked communication and computation, explores glitch and systemic bias, and extends these concepts to reversal of everything from identity to relationships. Kyle has been an adjunct professor at NYU's ITP, and a member of F.A.T. Lab, community manager for openFrameworks, and artist in residence at STUDIO for Creative Inquiry at Carnegie Mellon, as well as YCAM in Japan. His work is commissioned by and shown at exhibitions and festivals around the world, including: NTT ICC, Ars Electronica, Sonar/OFFF, Eyebeam, Anyang Public Art Project, Cinekid. He frequently leads workshops exploring computer vision and interaction.
In addition to the performances, please join us for a Workshop and an Artist Talk by select members of the creative team in the week leading up to the shows.
May 13-15, 2019: Machine Learning & Computer Vision Workshop
May 16, , 2019: discrete figures Artist Talk: Daito Manabe, Mikiko & Kyle McDonald
May 17-18, 2019: 'discrete figures 2019' Performances
