Texas A&M University · Fall 2026 · Section 603
| Course Number | CSCE 689 |
| Title | Generative AI for Computer Vision |
| Section | 603 |
| Time | MW 4:10 PM – 5:25 PM |
| Location | ZACH 310 |
| Credit Hours | 3 |
| Prerequisites | Graduate classification |
| Instructor | Zhengzhong Tu |
| Office | PETR 220 |
| Phone | 979-845-5904 |
| tzz@tamu.edu | |
| Office Hours | By appointment |
This graduate course provides an in-depth study of generative artificial intelligence with an emphasis on computer vision. Topics include deep learning foundations; convolutional neural networks; advanced architectures such as vision transformers; autoregressive models; autoencoders (AEs), variational autoencoders (VAEs) and generative adversarial networks (GANs); diffusion models and their applications; creative AI; video generative models; large language models (LLMs); vision-language models (VLMs); agentic LLM systems and applications; and trustworthiness of generative AI systems. The course combines instructor-led lectures with student presentations and culminates in group project presentations.
By the end of the course, students will be able to:
- Identify and describe standard generative modeling methods in language and vision.
- Compare major classes of generative models, including their strengths and limitations.
- Explain key motivations for generative modeling and the problems it is designed to address.
- Recognize and evaluate common assessment methodologies for generative models in language and vision.
- Apply generative modeling techniques to real-world tasks and applications.
- Conduct literature reviews, synthesize recent research findings, and formulate novel research directions.
| Week | Topics |
|---|---|
| 1 | Introduction and Course Overview |
| 2 | Deep Learning Basics; Building a Deep Network; Representative Models and Tasks |
| 3 | Advanced Models: Transformers and MLPs in Vision |
| 4 | Foundation Vision Models: Pre-training and Fine-tuning |
| 5 | Robustness in Vision: Image Enhancement, Domain Generalization |
| 6 | Robustness in Vision: Adversarial Robustness, Out-of-Distribution |
| 7 | Generative Models: Autoregressive Models, VAEs and GANs |
| 8 | Generative Models: Diffusion Models and Applications |
| 9 | Large Language Models (LLMs) and Multimodal LLMs |
| 10 | Special Topics: Fine-tuning and Adaptation of LLMs and VLMs |
| 11 | Special Topics: Trustworthiness in Foundation Models |
| 12 | Special Topics: LLM Reasoning, Planning, and Agents |
| 13 | Buffer |
| 14 | Final Project Presentations |
| 15 | Final Project Presentations |
| Component | Weight |
|---|---|
| Assignments | 30% |
| — Assignment 1 | 10% |
| — Assignment 2 | 10% |
| — Assignment 3 | 10% |
| Quizzes | 24% |
| — Quiz 1 | 8% |
| — Quiz 2 | 8% |
| — Quiz 3 | 8% |
| Class Participation | 6% |
| Final Project | 40% |
| — Milestone 1 Report (2-page proposal) | 5% |
| — Milestone 2 Report (4-page midterm) | 5% |
| — Final Presentation | 15% |
| — Final Report and Code Review | 15% |
There is no final exam.
Grading scale: A = 90–100 · B = 80–89 · C = 70–79 · D = 60–69 · F = <60
Any academic misconduct in this course will result in a grade of F.
Assignments (30%). Three hands-on assignments reinforce the concepts covered in lectures: (1) deep learning foundations and convolutional neural networks, (2) advanced architectures (e.g., vision transformers), foundation vision models, and robustness, and (3) generative models and large multimodal models. Each assignment requires a written report and a code submission. Assignment 1 is due by end of Week 4, Assignment 2 by end of Week 8, and Assignment 3 by end of Week 12. Each is weighted 10%.
Quizzes (24%). Three in-class quizzes cover lecture material, assigned readings, and conceptual understanding of generative AI for computer vision. Each quiz is closed-book and weighted 8%. Quiz 1 is held by end of Week 5, Quiz 2 by end of Week 9, and Quiz 3 by end of Week 13.
Class Participation (6%). Students are expected to actively engage in class discussions, in-class exercises, and Q&A. Participation is assessed throughout the semester (Weeks 1–15) based on attendance, in-class engagement, and contributions to discussions and paper review activities.
Final Project (40%). Students work in small teams on an end-to-end research project related to generative AI for computer vision: identify a research question, conduct a literature review, propose and implement a methodology, run experiments, and analyze results. Four graded components:
- Milestone 1 — Proposal Report (5%): A 2-page report defining the research problem, motivation, related work, proposed approach, and expected outcomes. Due by end of Week 6.
- Milestone 2 — Midterm Report (5%): A 4-page report summarizing preliminary results, refined methodology, and remaining work. Due by end of Week 10.
- Final Presentation (15%): An in-class team presentation of project objectives, methods, results, and findings, followed by Q&A. Held during Weeks 14–15.
- Final Report and Code Review (15%): A complete final written report with code submission describing methodology, experiments, results, analysis, and discussion. Due by end of Week 15.
Each additional late day incurs a 10% penalty, no exceptions:
| Late | Penalty |
|---|---|
| 1 day | 10% |
| 2 days | 20% |
| 3 days | 30% |
| 4 days | 50% |
| 5+ days | 100% |
Work submitted as makeup work for an excused absence is not considered late work and is exempt from this policy (Student Rule 7). Students are responsible for informing the instructor in a timely manner about excused absences; the instructor will then work with the student to catch up on submissions, including missed group class participation.
No textbook is required. Lecture slides, videos, and other materials provided by the instructor serve as the primary reference. Recommended supplementary texts:
- Foundations of Computer Vision — Antonio Torralba, Philip Isola, William T. Freeman (2024)
- Deep Learning — Ian Goodfellow, Yoshua Bengio, Aaron Courville (2016)
- Computer Vision: Algorithms and Applications — Richard Szeliski (2010)
- Dive into Deep Learning — Aston Zhang, Zack Lipton, Mu Li, Alex Smola (2019)
- The Little Book of Deep Learning — François Fleuret (2024)
- Probabilistic Machine Learning: An Introduction — Kevin Murphy (2022)
Attendance and Makeup Work
The university views class attendance and participation as an individual student responsibility. Students are expected to attend class and to complete all assignments.
Students will be excused from attending class on the day of a graded activity, or when attendance contributes to a student's grade, for the reasons stated in Student Rule 7 or other reason deemed appropriate by the instructor.
Absences related to Title IX of the Education Amendments of 1972 may necessitate a period of more than 30 days for makeup work, and the timeframe for makeup work should be agreed upon by the student and instructor (Student Rule 7, Section 7.4.1).
"The instructor is under no obligation to provide an opportunity for the student to make up work missed because of an unexcused absence" (Student Rule 7, Section 7.4.2).
Students who request an excused absence are expected to uphold the Aggie Honor Code and Student Conduct Code (see Student Rule 24).
Please refer to Student Rule 7 in its entirety for information about excused absences and makeup work, including definitions, documentation, and timelines.
Academic Integrity
"An Aggie does not lie, cheat or steal, or tolerate those who do."
"Texas A&M University students are responsible for authenticating all work submitted to an instructor. If asked, students must be able to produce proof that the item submitted is indeed the work of that student. Students must keep appropriate records at all times. The inability to authenticate one's work, should the instructor request it, may be sufficient grounds to initiate an academic misconduct case" (Section 20.1.2.3, Student Rule 20).
You can learn more about the Aggie Honor System Office Rules and Procedures, academic integrity, and your rights and responsibilities at aggiehonor.tamu.edu.
Notice of Nondiscrimination
Texas A&M University is committed to providing safe and non-discriminatory learning, living, and work environments for all members of the University community. The University provides equal opportunity to all employees, students, applicants for employment or admission, and the public regardless of race, color, sex (including pregnancy and related conditions), religion, national origin, age, disability, genetic information, or veteran status. Texas A&M University will promptly, thoroughly, and fairly investigate and resolve all complaints of discrimination, harassment (including sexual harassment), complicity and related retaliation based on a protected class in accordance with System Regulation 08.01.01, University Rule 08.01.01.M1, Standard Administrative Procedure (SAP) 08.01.01.M1.01, and applicable federal and state laws. In accordance with Title IX and its implementing regulations, Texas A&M does not discriminate on the basis of sex in any educational program or activity, including admissions and employment. The following person has been designated to handle inquiries and complaints regarding the non-discrimination policies: Jennifer M. Smith, TAMU Associate VP & Title IX Coordinator at YMCA Ste 108, College Station, TX 77843, 979-458-8407, or email civilrights@tamu.edu. For other reporting options, visit https://ocrcas.ed.gov/contact-ocr to locate the address and phone number of the office that serves your area, or call 1-800-421-3481.
Civil Rights, Free Speech, and Title IX
Texas A&M University is committed to fostering a learning environment that is safe and productive for all. University policies and federal and state laws prohibit discrimination and harassment based on an individual's race, color, sex (including pregnancy and related conditions), religion, national origin, age, disability, genetic information, veteran status, or any other legally protected characteristic. This includes forms of sex-based violence, such as sexual assault, sexual harassment, sexual exploitation, dating/domestic violence, and stalking.
Students can report discrimination/harassment, access supportive resources, or learn more about their options for resolving complaints on the University's Civil Rights & Title IX webpage.
Students should be aware that all university employees (except medical or mental health providers) are mandatory reporters, which means that if they observe, experience, or become aware of an incident that they reasonably believe to be discrimination/harassment alleged to have been committed by or against a person who was a student or employee at the time of the incident, the employee must report the incident to the university.
Americans with Disabilities Act (ADA)
Texas A&M University is committed to providing equitable access to learning opportunities for all students. If you experience barriers to your education due to a disability or think you may have a disability, please contact the Disability Resources office on your campus. Disabilities may include, but are not limited to, attentional, learning, mental health, sensory, physical, or chronic health conditions. All students are encouraged to discuss their disability-related needs with Disability Resources and their instructors as soon as possible.
Students at Texas A&M University, College Station should contact Disability Resources at (979) 845-1637 or disability@tamu.edu.
If you are experiencing difficulties with your approved accommodations, contact the office responsible for approving your accommodations or the Texas A&M ADA Coordinator, Julie Kuder, at ADA.Coordinator@tamu.edu or (979) 458-8407.
Pregnancy Accommodations
Texas A&M provides reasonable accommodations to students due to pregnancy and/or related conditions, such as childbirth, recovery, and lactation. Students should contact the University's Pregnancy Coordinator as soon as they become aware of the need for accommodation. Depending on the circumstances, accommodations could include extended time to complete assignments or exams, changes in course sequence, or modifications to the physical classroom environment. Texas A&M will also allow a voluntary leave of absence, ensure the availability of lactation space, and maintain grievance procedures to provide for the prompt and equitable resolution of complaints of sex discrimination. For information regarding pregnancy accommodations, email TIX.Pregnancy@tamu.edu.
Mental Health and Wellness
Texas A&M University recognizes that mental health and wellness are critical factors influencing a student's academic success and overall wellbeing. Students are encouraged to engage in healthy self-care practices by utilizing the resources and services available through University Health Services on its mental health webpage. The TELUS Health Student Support app provides access to professional counseling in multiple languages anytime, anywhere by phone or chat, and the 988 Suicide & Crisis Lifeline offers 24-hour emergency support at 988 or 988lifeline.org.
Students at College Station needing a listening ear can contact University Health Services at (979) 458-4584. 24-hour emergency help is also available through the 988 Suicide & Crisis Lifeline (988) or at 988lifeline.org.
Family Educational Rights and Privacy Act (FERPA)
FERPA is a federal law designed to protect the privacy of educational records by limiting access to these records, to establish the right of students to inspect and review their educational records, and to provide guidelines for the correction of inaccurate and misleading data through informal and formal hearings. Currently enrolled students wishing to withhold any or all directory information items can do so within howdy.tamu.edu using the Directory Information Withholding Form. The complete FERPA Notice to Students and the student records policy is available on the Office of the Registrar webpage.
Items that can never be identified as public information are a student's social security number, citizenship, gender, grades, GPR, or class schedule. All efforts will be made in this class to protect your privacy and to ensure confidential treatment of information associated with or generated by your participation in the class.
Directory items include name, UIN, local address, permanent address, email address, local telephone number, permanent telephone number, dates of attendance, program of study (college, major, campus), classification, previous institutions attended, degrees, honors and awards received, participation in officially recognized activities and sports, medical residence location, and medical residence specialization.
Free Speech and Civil Discourse
Texas A&M recognizes that the pursuit of truth through open and robust discourse is critical to academic inquiry. However, as a community of scholars, the university has an aspirational expectation that such discourse will be conducted in accordance with Aggie Core Values. In this "marketplace of ideas," we encourage civil dialogue creating an environment that allows individuals to express their ideas and to have their ideas challenged in respectful and responsible ways. Students can learn more about Freedom of Expression and Free Speech on the University's website about the First Amendment.