|
Jason J. Corso
Teaching List
Contents Links
Michigan Courses
Tutorials
Buffalo Courses
|
This page gives a list and brief description of the courses I now teach
(and have taught in past semesters).
| |
Winter 2017
|
| EECS 542: |
Advanced Topics In Computer Vision
|
| URL: |
/~jjcorso/t/542W17/ |
| Description: |
The course will focus on learning structured representations and embeddings for high-level problems in computer vision. Approaches for structured prediction, deep learning, and dictionary learning will be covered, all with an emphasis on modeling certain classes of structure, such as affine invariance and sparsity. The course will be highly interactive with a mix of readings, homeworks, quizzes and a course project.
|
| |
Fall 2016
|
| EECS 504: |
Foundations of Computer Vision
|
| URL: |
/~jjcorso/t/504F16/ |
| Description: |
Computer Vision seeks to extract useful information from images. This
course covers the foundations of computer vision. It emphasizes
computer vision as a search for visual invariants and computer vision
as mathematical modeling. Foundational representations of images and
image content will be discussed. Cross-cutting problems of
reduction (e.g., feature extraction, segmentation), estimation (e.g.,
image alignment, camera calibration), and matching will be concretely
defined and elaborated through many real examples from modern computer
vision.
|
| |
Winter 2016
|
| EECS 542: |
Advanced Topics In Computer Vision
|
| URL: |
/~jjcorso/t/542W16/ |
| Description: |
The course will focus around learning structured representations and embeddings for numerous problems in computer vision. Approaches for learning from unimodal and multimodal data will be covered and include topics from sparse coding, convolutional neural networks and others. Multimodal problems will include topics at the boundary of vision and language, vision and speech, and vision and robotics. The course will be a mix of lecture, student presentation, and discussion; students will complete some programming assignments, paper reviews and presentations, and a project. Prior experience from EECS 442 or EECS 5XX (Foundations of Computer Vision) expected.
|
| |
Fall 2015
|
| EECS 598-01: |
Foundations of Computer Vision
|
| URL: |
/~jjcorso/t/598F15/ |
| Description: |
Computer Vision seeks to extract useful information from images. This
course covers the foundations of computer vision. It emphasizes
computer vision as a search for visual invariants and computer vision
as mathematical modeling. Foundational representations of images and
image content will be discussed. Cross-cutting problems of
reduction (e.g., feature extraction, segmentation), estimation (e.g.,
image alignment, camera calibration), and matching will be concretely
defined and elaborated through many real examples from modern computer
vision.
|
| |
Winter 2015
|
| EECS 598-04: |
Probabilistic Graphical Models in Vision and Beyond
|
| URL: |
/~jjcorso/t/598W15/ |
| Description: |
Probabilistic graphical models have emerged as a powerful formalism for leveraging principled probability theory and discrete structured data representations to model large-scale problems involving hundreds or even thousands of often high-dimensional variables. Computer vision provides a set of such large-scale, concrete problems well-suited for probabilistic graphical models. The course will cover probabilistic graphical models in detail starting from the basics and pushing through contemporary results. There will be an emphasis on driving problem formulations from computer vision but our coverage will be broad; connections to other application areas will be discussed when plausible.
|
| |
Fall 2014
|
| EECS 598-08: |
Foundations of Computer Vision
|
| URL: |
/~jjcorso/t/598F14/ |
| Description: |
Computer Vision seeks to extract useful information from images. This course begins the fundamentals of image formation and then organizes the remaining material according to the class of information to be extracted. We will cover early processes, such as basic features, edges and contours; motion tracking, including optical flow and filtering; shape primarily from a binocular 3D reconstruction point of view; and both object and action detection and recognition. The course has been designed to present an introduction to computer vision targeted to graduate students. The course will balance theory and application both in lectures and assignments.
|
| |
CVPR 2014 Tutorial
|
| CVPR 2014: |
Tutorial on Video Segmentation
|
| URL: |
/~jjcorso/t/CVPR14_videoseg/ |
| Description: |
In recent years, segmentation has emerged as a plausible first step in
early video processing of unconstrained videos, without needing to
make an assumption of a static background as earlier methods have.
Video segmentation and over-segmentation, or more commonly
supervoxel extraction, is a complementary early video
processing step to the more traditional feature extraction, such as
STIP and trajectories, and it extends the long history of image
segmentation methods. This tutorial will survey and present the
important models and algorithms for video segmentation. We will cover
direct extensions of image segmentation methods through video-specific
spatiotemporal and streaming methods. In addition to core
methodological elements, the tutorial will also cover benchmark and
evaluation of video segmentation as well as applications of video
segmentation. Participants will be introduced to the details of these
methods not only through traditional slide presentations but also
example implementations through the LIBSVX library.
|
| |
Spring 2014
|
| CSE 704: |
Readings in Joint Visual, Lingual and Physical Models and Inference Algorithms
|
| URL: |
http://www.cse.buffalo.edu/~jcorso/t/2014S_SEM/ |
| Description: |
This seminar will study modeling and inference in the case that multimodal data is available. The data situations of focus on vision and language, but others will be considered, such as action (physical motion), audition, etc. The seminar will focus on reading and discussing topic-relevant research papers.
|
| |
Spring 2013
|
| CSE 455/555: |
Introduction to Pattern Recognition |
| URL: |
http://www.cse.buffalo.edu/~jcorso/t/2013S_555/ |
| Description: |
Foundations of pattern recognition algorithms and machines, including statistical and structural methods. Data structures for pattern representation, feature discovery and selection, classification vs. description, parametric and non-parametric classification, supervised and unsupervised learning, use of contextual evidence, clustering, recognition with strings, and small sample-size problems.
|
| |
Fall 2012
|
| CSE 672: |
Bayesian Vision |
| URL: |
http://www.cse.buffalo.edu/~jcorso/t/2012F_672/ |
| Description: |
The course takes an in-depth look at various Bayesian methods in
computer and medical vision. Through the language of Bayesian
inference, the course will present a coherent view of the approaches
to various key problems such as detecting objects in images,
segmenting object boundaries, and recognizing objects. The course is
roughly partitioned into two halves: modeling and inference. In the
first half, it will cover both classical models such as weak membrane
models and Markov random fields as well as more recent models such as
conditional random fields, latent Dirichlet allocation, and topic
models. In the second half, it will focus on inference algorithms.
Methods include PDE boundary evolution algorithms such as region
competition, discrete optimization methods such as graph-cuts and
graph-shifts, and stochastic optimization methods such as data-driven
Markov chain Monte Carlo. An emphasis will be placed on both the
theoretical aspects of this field as well as the practical application
of the models and inference algorithms.
|
| |
Spring 2012
|
| CSE 455/555: |
Introduction to Pattern Recognition |
| URL: |
http://www.cse.buffalo.edu/~jcorso/t/2012S_555/ |
| Description: |
Foundations of pattern recognition algorithms and machines, including statistical and structural methods. Data structures for pattern representation, feature discovery and selection, classification vs. description, parametric and non-parametric classification, supervised and unsupervised learning, use of contextual evidence, clustering, recognition with strings, and small sample-size problems.
|
| |
Fall 2011
|
| CSE 734: |
Seminar: Readings in Computer Vision and Machine Learning
|
| URL: |
http://www.cse.buffalo.edu/~jcorso/t/2011F_SEM/ |
| Description: |
This is a seminar course in advanced topics in computer vision and machine learning.
We will read and discuss papers on this topic throughout the semester, with the students primarily in charge of leading the discussions.
|
| |
Spring 2011
|
| CSE 455/555: |
Introduction to Pattern Recognition |
| URL: |
http://www.cse.buffalo.edu/~jcorso/t/2011S_555/ |
| Description: |
Foundations of pattern recognition algorithms and machines, including statistical and structural methods. Data structures for pattern representation, feature discovery and selection, classification vs. description, parametric and non-parametric classification, supervised and unsupervised learning, use of contextual evidence, clustering, recognition with strings, and small sample-size problems.
|
| |
Fall 2010
|
| CSE 672: |
Bayesian Vision |
| URL: |
http://www.cse.buffalo.edu/~jcorso/t/2010F_672/ |
| Description: |
The course takes an in-depth look at various Bayesian methods in
computer and medical vision. Through the language of Bayesian
inference, the course will present a coherent view of the approaches
to various key problems such as detecting objects in images,
segmenting object boundaries, and recognizing objects. The course is
roughly partitioned into two halves: modeling and inference. In the
first half, it will cover both classical models such as weak membrane
models and Markov random fields as well as more recent models such as
conditional random fields, latent Dirichlet allocation, and topic
models. In the second half, it will focus on inference algorithms.
Methods include PDE boundary evolution algorithms such as region
competition, discrete optimization methods such as graph-cuts and
graph-shifts, and stochastic optimization methods such as data-driven
Markov chain Monte Carlo. An emphasis will be placed on both the
theoretical aspects of this field as well as the practical application
of the models and inference algorithms.
|
| CSE 705: |
Vision Seminar (Spatiotemporal Video Analysis) |
| URL: |
http://www.cse.buffalo.edu/~jcorso/t/2010F_SEM/ |
| Description: |
This is a seminar course covering topics in spatiotemporal video analysis.
We will read and discuss papers on this topic throughout the semester, with the students primarily in charge of leading the discussions.
|
| UE 141 DD: |
Discovery Seminar in Computer Vision |
| URL: |
http://www.cse.buffalo.edu/~jcorso/t/2010F_DIS/ |
| Description: |
This seminar will explore the intriguing and often misunderstood field of Computer Vision---automatic computer interpretation of visual data. We will read about and watch various portrayals of core Computer Vision problems in popular culture, such as Robocop and Minority Report. In class, we will discuss these core research problems in light of the readings and videos to allow the students to build a understanding of where the exciting field of Computer Vision is now and where it is going in the future. We will rely heavily on the Computer Vision Fact and Fiction project at UCSD.
|
| |
Spring 2010
|
| CSE 555: |
Introduction to Pattern Recognition |
| URL: |
http://www.cse.buffalo.edu/~jcorso/t/2010S_555/ |
| Description: |
Foundations of pattern recognition algorithms and machines, including statistical and structural methods. Data structures for pattern representation, feature discovery and selection, classification vs. description, parametric and non-parametric classification, supervised and unsupervised learning, use of contextual evidence, clustering, recognition with strings, and small sample-size problems.
|
| |
Fall 2009
|
| CSE 642: |
Techniques in AI: Vision for HCI |
| URL: |
http://www.cse.buffalo.edu/~jcorso/t/2009F_642/ |
| Description: |
The promise of computer vision for enabling natural human-machine interfaces is great: vision-based interfaces would allow unencumbered, large-scale spatial motion; they could make use of hand gestures, movements, or other similar natural input means; and video itself is passive, cheap, and nearly ubiquitous. In the simplest case, tracked hand motion and gesture recognition could replace the mouse in traditional applications, but, computer vision offers the additional possibility of using both hands simultaneously, using the face, incorporating multiple users concurrently, etc. In this course, we will develop these ideas from both a theoretical and a practical perspective. From the theoretical side, we will cover ideas ranging from interaction paradigms suitable for vision-based interfaces to mathematical models for tracking (e.g., particle filtering), modeling high-dimensional articulated objects, and modeling a grammar of interaction, as well as algorithms for rapid and real-time inference suitable for interaction scenarios. From the practical side, we will each build (in pairs) an actual working vision-based interactive system. Each project must "close the loop" and be integrated directly into an interactive computer system (e.g., sort photos on the screen by grabbing them with each hand and moving them around). During the semester, very practical-minded topics such as interactive system design and architecture, debugging programs that process high-dimensional video data, and program optimization will be discussed alongside the underlying computer vision theory.
|
| CSE 702: |
Seminar in Image Semantics |
| URL: |
http://www.cse.buffalo.edu/~jcorso/t/2009F_702/ |
| Description: |
This course will explore the topic of semantics in image and video analysis. We will read and discuss papers on this topic throughout the semester, with the students primarily in charge of leading the discussions.
|
| |
Spring 2009
|
| CSE 555: |
Introduction to Pattern Recognition |
| URL: |
http://www.cse.buffalo.edu/~jcorso/t/2009S_555/ |
| Description: |
Foundations of pattern recognition algorithms and machines, including statistical and structural methods. Data structures for pattern representation, feature discovery and selection, classification vs. description, parametric and non-parametric classification, supervised and unsupervised learning, use of contextual evidence, clustering, recognition with strings, and small sample-size problems.
|
| |
Fall 2008
|
| CSE 702: |
Seminar in Pattern Theory |
| URL: |
http://www.cse.buffalo.edu/~jcorso/t/2008F_702/ |
| Description: |
This seminar will focus on Grenander's Pattern Theory from a practical, contemporary perspective. Pattern Theory is the study of patterns from a representational perspective rather than a recognition one. Miller and Grenander write "Pattern theory attempts to provide an algebraic framework for describing patterns as structures regulated by rules, essentially a finite number of both local and global combinatory operations. Pattern theory takes a compositional view of the world, building more and more complex structures starting from simple ones. The basic rules for combining and building complex patterns from simpler ones are encoded via graphs and rules on transformations of these graphs." We will explore various theoretical aspects of modern pattern theory (e.g., probabilistic graphical models, grammars, matrix groups, information measures, manifolds, Markov processing and sampling) in the context of practical problems in computer vision and medical imaging. Students will be required to give one or two (depending on seminar size) prepared lectures during the semesters. Grading is S/U; letter grading is available as an option and requires a project.
|
| |
Spring 2008
|
| CSE 672: |
Vision as Bayesian Inference |
| URL: |
http://www.cse.buffalo.edu/~jcorso/t/2008spring_vbi/ |
| Description: |
The course takes an in-depth look at various Bayesian methods in
computer and medical vision. Through the language of Bayesian
inference, the course will present a coherent view of the approaches
to various key problems such as detecting objects in images,
segmenting object boundaries, and recognizing objects. The course is
roughly partitioned into two halves: modeling and inference. In the
first half, it will cover both classical models such as weak membrane
models and Markov random fields as well as more recent models such as
conditional random fields, latent Dirichlet allocation, and topic
models. In the second half, it will focus on inference algorithms.
Methods include PDE boundary evolution algorithms such as region
competition, discrete optimization methods such as graph-cuts and
graph-shifts, and stochastic optimization methods such as data-driven
Markov chain Monte Carlo. An emphasis will be placed on both the
theoretical aspects of this field as well as the practical application
of the models and inference algorithms.
|
| |
Fall 2007
|
| CSE 702: |
Seminar in Medical Image Segmentation |
| URL: |
http://www.cse.buffalo.edu/~jcorso/t/2007fall_smis/ |
| Description: |
The seminar will survey the literature in medical image segmentation.
Topics include knowledge-based heuristics, voxel-based statistics,
contour evolution, hierarchical modeling, and learning-based
approaches. We will focus on constructing a complete taxonomy of
approaches in this area. Students will be required to make one paper
presentation and do a project to explore a method, which can be new
research, in detail. Familiarity with vision and medical image
computing is suggested but not required.
|
While a postdoc at UCLA.
| |
Spring 2006
|
| BIOMED223C |
Programming Lab in Medical Informatics III |
| Topic: |
Object-Oriented Methods in Software
Engineering for Medical Informatics |
| URL: |
/~jjcorso/t/biomed223c-spring06/
|
| Description: |
The course is designed to expose the students to both relevant topics
in medical informatics and the process of developing these topics into
large software systems. The topic of emphasis for this quarter will be
pattern classification. As a group, we will design the software
framework necessary for a comprehensive pattern classification system.
Collectively, the students will implement the designed framework.
Individually, each student will implement a particular pattern
classification algorithm in this framework. This project will be
developed through the duration of the quarter as new topics in both
software engineering and pattern classification are learned; the
result will be a practical and complete system that the students can
take with them into their future research.
|
|
