2026f


Instructor
Teaching Assistante


MAT255 Techniques, History & Aesthetics of the Computational Photographic Image
George Legrady | vislab.mat.ucsb.edu
Shaw (Yiran) Xiao

Course materials are protected by US Copyright laws and by University policy. Contents of this course may not be reproduced, distributed, or displayed without my express prior written consent.

Tues-Thurs 3:30-5:20pm (some lectures may be online) otherwise Lab 2611, 2nd flr, Elings Hall
Course Information
This course introduces practical skills in ComfyUI (comfy.org), an open-source visual AI platform that uses a modular, node-based interface to produce images, audio, and video. Alongside hands-on technique, students will engage critically with generative AI through discussions, presentations, and research — examining its conceptual, aesthetic, and cultural implications. ComfyUI's 60,000+ community-built Python nodes allow for precise, flexible creative control.
Course Workload

 . Sign the Legal agreement
. Student Forum: Where you post your work
. ComfyUI Workspace: the workspace for workflow)
. Attendance and participation at both in class meetings and zoom lectures
. Final presentation pdf documentation of either a research paper basedon a project
Resources

 Campus VPN | Access to Dell Pro Max
ComfyUI Summary | Tutorial-Documentation | GitHub | Examples
09.24
 Course Introduction
AI & Image Knowledge | Human / Machine Art
09.29


10.01
 		Text To Image | Prompt Engineering | Examples | Lakoff Metaphors
Text prompt generates an image. Prompt engineering, flux, checkpoints, sampler, denoise, seed.

Text Prompt Presentations
10.06

10.08
 Img2Img | Diffusion explainer

Image-to-image Presentations
10.13

10.15
 XY Plot

XY Plot Presentations
10.20

10.22
 Style Transfer

Style Transfer Presentations
10.27

10.29
 LORA [1] [2]| Flux1.dev |

Lora Presentations
11.03

11.05
 ControlNet: Input form determines output form, can also provide image depth analysis

ControlNet Presentations
11.10


11.12
 Diverse Image Processing: Mask, noise, image analysis, etc. Expands on Photoshop and Matlab processes

Image Processing Presentations
11.17

11.19 		AnimateDiff: Animate a still image. Interative generation with varying seeds

AnimateDiff Presentation
11.24

11.26
 Runway | Gemini Google | Stable Video Diffusion | Seedance |Wan

Video Presentations
12.01

12.03
 Time-Based Audio | Rave

Audio Presentations
12.08

12.10
 Final Presentations

Final Presentations
Misc Reference Links
Lev Manovich
Payton Croskey Suggestion
Charlie Price Suggestion
Fani Harid
Noam Elcott
Fabian Offert
Katrina Sluis
A questionnaire on Art and Machine Learning
Artificial Aesthetics: A Critical Guide to AI: Media and Design
Mapping the Mind of a Large Language Model [research paper] |  Models All the Way Down
The Weird and the Eerie (Marc Fisher) | Unheimliche / Uncanny
TED talk | UC Berkeley | Interview (start around 13:00)
AI Imaging
The Method of Critical AI Sudies, A Propaedeutic
Image Generation in the Google Research Lab
Student Projects
Final Projects Documentations