Intelligent Visual Computing: A Neural Network Approach
Zoom Meeting time:
Tuesdays + Thursdays 1-2:15pm ET (check moodle for zoom link)
Remote Instructor office hours:
Tuesdays + Thursdays 2:15-3pm ET
Tuesdays + Thursdays 10:30-11pm ET (for students in different time zones)
Instuctor: Evangelos Kalogerakis (
Location: FULLY REMOTE CLASS ON ZOOM (check moodle for zoom link)
Reemote TA office hours:
TBA (check moodle for zoom link)
This course covers deep learning and AI methods for 3D computer graphics, 3D computer vision, and mixed reality. It introduces students to a wide range of problems related to the processing and creation of visual content: 3D models, scenes, animations, images, and video. For each problem, the course describes state-of-the-art learning-based solutions based on recent research findings. Topics include 3D shape classification, segmentation, correspondences, reconstruction, synthesis, deformations, texture synthesis, animation, and neural rendering. Students review research papers, and complete 5 programming assignments in Python/PyTorch. Students taking the 674 section will also work on an open-ended course project related to the topics covered in this course.
CMPSCI graduate students: there are no prerequisites for CMPSCI graduate students, although familiarity with probability, statistics, linear algebra, undergraduate computer graphics and machine learning is essential!
Undergrads and others: the course has the following prerequisites:
- Introduction to Algorithms (311 or equivalent) with grade B or better
- Artificial intelligence (383 or equivalent) with grade B or better
- Introduction to Computer Graphics (373, 473 or equivalent) with grade B or better
This course counts as a CS Elective toward the CMPSCI major (BA/BS) and can satisfy a second or third AI core requirement in the CMPSCI PhD and MSc program. It offers 3 credits.
Remote Class planLecture material will be pre-recorded and available asynchronously. Scheduled class time will be used for additional (synchronous) material and live discussion (attendance is optional, yet encouraged). These zoom meetings will also be recorded. There will be scheduled office hours with the instructor and the TAs, including additional instructor office hours for students in different time zones. Team work can be done asynchronously. Zoom links, lecture notes, recorded videos, programming resources, and assignments are posted on Moodle. Piazza will be used for asynchronous Q & A with the instructor and TAs.
Week-by-week outline (tentative)
Week 1: Image and Shape Representations, Basics of Classification & Regression, Intro to Neural Networks
Week 2: 3D Deep Learning: Multi-view networks, 3D Shape Segmentation and Correspondences
Week 3: 3D Deep Learning: Volumetric networks, Implicit field networks
Week 4: 3D Deep Learning: Point-based networks, mesh-based networks
Week 5: 3D Generative Networks
Week 6: 3D Generative Networks (cont'd)
Week 7: 4D Deep Learning
Week 8: Neural Deformations, Neural Rigging
Week 9: Neural Animation
Week 10: Neural Rendering
Week 11: Neural Rendering (cont'd)
Week 12-13: Paper presentations / Project presentations
- Complete five assignments
- Research paper presentation
- Research paper reaction reports
- Complete five assignments
- A term project + presentation. This should be an implementation of a deep learning algorithm related to the topics of the course. The result of the project should be a prototype with some preliminary results, or could lead to a submission of a research paper in the future.
- Research paper reaction reports
Marking scheme (tentative)
80% assignments, reaction reports
20% research paper presentation
70% assignments, reaction reports
30% project + presentation
Scores to letter conversions:
Note: A course grade below C is not a passing grade for graduate students.
Should you take 574 or 674?
Both 574 and 674 sections meet at the same time, same location, and have access to the same lectures and course material. The programming assignments and reaction report requirements are the same. The difference is that 574 requires students to present an existing research paper towards the end of the course (~15 min presentation), while 674 requires students to develop and implement (on their own) a deep learning method related to the topics of the course as a final course project. The project can be an implementation of a (small-scale) research idea/prototype, or could lead to a submission of a research paper in the future. The project plan must be first discussed with the instructor (soon after the middle of the semester). Students also present their project work towards the end of the course. As a result, 674 requires more effort and is more targeted for students interested in graphics+vision research.
Accommodation StatementThe University of Massachusetts Amherst is committed to providing an equal educational opportunity for all students. If you have a documented physical, psychological, or learning disability on file with Disability Services (DS), you may be eligible for reasonable academic accommodations to help you succeed in this course. If you have a documented disability that requires an accommodation, please notify me within the first two weeks of the semester so that we may make appropriate arrangements.
Academic Honesty StatementSince the integrity of the academic enterprise of any institution of higher education requires honesty in scholarship and research, academic honesty is required of all students at the University of Massachusetts Amherst. Academic dishonesty is prohibited in all programs of the University. Academic dishonesty includes but is not limited to: cheating, fabrication, plagiarism, and facilitating dishonesty. Appropriate sanctions may be imposed on any student who has committed an act of academic dishonesty. Instructors should take reasonable steps to address academic misconduct. Any person who has reason to believe that a student has committed academic dishonesty should bring such information to the attention of the appropriate course instructor as soon as possible. Instances of academic dishonesty not related to a specific course should be brought to the attention of the appropriate department Head or Chair. Since students are expected to be familiar with this policy and the commonly accepted standards of academic integrity, ignorance of such standards is not normally sufficient evidence of lack of intent (http://www.umass.edu/dean_students/codeofconduct/acadhonesty/).