Class time: Tuesday and Thursday, 1:50PM - 2:45PM
Classroom: University Hall 090
Course website: https://sites.google.com/view/osu-cse5539-mlcv
Instructor: Prof. Wei-Lun (Harry) Chao
Email: chao.209@osu.edu
Office hour: Thursday 3:00PM - 4:00PM or by appointment
Course abstract: This research course will cover various machine learning (ML) topics and their applications to computer vision (CV). Specifically, we will study ML topics such as transfer learning, domain adaptation, meta-learning, generative models, imbalanced learning, and semi-supervised learning. We will then study how these topics and corresponding techniques are applied to CV problems such as object classification, detection, segmentation, and visual question answering. The format of the class will be a mix of lectures and research paper presentations. Students who participate in this class are expected to be self-motivated graduate or senior undergraduate students.
Pre-requisites: Students are expected to (self) learn how to use deep learning software (e.g., Tensorflow and Pytorch). Students are expected to have a strong interest in machine learning and computer vision and have taken courses in artificial intelligence/machine learning (5522, 5523, or 5526) or computer vision (5524).
Syllabus: link
Grading:
Quiz & participation: 10%
Participation includes asking questions, discussion in the lectures
Paper presentation & survey: 35% (20%: presentation + 15%: report)
The presentation is graded based on efforts and the clearness in presenting ideas of the papers. The survey report is graded based on efforts and the clearness and how well you organize the papers that you read. The report should include sections like introduction, background, approach, experiments, etc.
Final project (1-4 people): 55% (15%: first+ 25%: presentation (results)+ 15%: report)
First presentation: 5% for oral presentation, 10% for written first report (1-2 pages).
Required Textbook: No required textbook
Suggested references:
Shai Shalev-Shwartz and Shai Ben-David, Understanding machine learning: From theory to algorithms. Cambridge university press, 2014.
Christopher M Bishop, Pattern recognition and machine learning. springer, 2006.
Ian Goodfellow, Yoshua Bengio, and Aaron Courville, Deep learning. MIT press, 2016.
Tutorials & workshop talks in CVPR, ICCV, ECCV, ICML, and NeurIPS
Survey report:
NeurIPS 2019 Latex style: https://nips.cc/Conferences/2019/CallForPapers
Extend your paper presentation from 3 papers to 6-10 papers
At least 8 pages, excluding reference (Introduction, an overview of the background, descriptions of some key algorithms and their concepts, and important experimental results and findings)
Due day: 3 weeks after your presentation. Later reports will lead to 20% deduction on your report score.
The survey report is graded based on efforts and the clearness and how well you organize the papers that you read. The report should include sections like introduction, background, approach, experiments, etc.
Example:
https://link.springer.com/article/10.1007/s10994-019-05855-6
Final project (comprehensive survey, individual project, challenge: 1-4 people):
Example challenges:
Other suggestions from NeurIPS, ICML, ICLR, ICCV, ECCV, CVPR workshops
Week schedule:
Slides: (slides are moved to Carmen Files)
Generative models: Stanford_CS231n, Shakir_Mohamed_DLSS_2016 (slides, video), Shakir_Mohamed's_other_tutorials