录取要求

MSCV Program: Recommended Skill Set
Mathematics

Linear Algebra (Inversion, Eigen Decomposition, Null Space)
Linear Differential Equations (Matrix Algebra, Matrix manipulation)
Calculus (Derivatives, Gradients, Chain Rule)
Probability and Statistics
Numerical Integration
Fourier Analysis
Optimization (convex optimization, Levenberg-Marquardt, sparse optimization)
Programming knowledge

Program constructs (Sequencing, Selection, Iteration and Recursion)
Data structures (Arrays, Lists, Pointers)
Matlab, C/C++, Python
OpenCV, VLfeat, Matlab Image processing toolbox
GPU (CUDA, OpenCL)
Software

Productivity software (MS office, iWork, OpenOffice)
Windows, MacOS, Linux
LaTeX
Photoshop (or equivalent)

申请材料清单

Admission & Application Requirements

Undergraduate (B.S. or equivalent) in engineering, computer science or applied mathematics REQUIRED

Standardized Test Completion

General GRE REQUIRED
TOEFL / IELTS (foreign students only) REQUIRED
Application Materials

Résumé
Statement of Purpose (1 to 2 pages)
Undergraduate/Graduate (as applicable) Transcripts
Letters of Recommendation (3 REQUIRED)
The next application period will be approximately from early September to mid-December 2018; accepted & enrolled students would begin in Fall 2019. Applications should be submitted using the online application form, linked below. More information is available via the School of Computer Science’s Admissions FAQ

专业介绍

Computer vision is the study of acquiring and interpreting visual imagery. With the exponential growth of digital images and videos captured by cameras the automated understanding of our visual world has never been more important. The field is advancing rapidly and its applications continue to expand in areas of great societal value. As computer vision shifts from research to development, there is a critical need for developers with expertise in this field. To meet the growing demand, The Robotics Institute has developed a 16 month (three semesters plus summer) professional Master’s program in Computer Vision (MSCV).

The goals of the MSCV program are to:

Provide a robust set of courses encompassing current and emerging state of the art computer vision topics that will prepare students for careers in this field.
Facilitate hands-on experience on real research and development projects addressing current applications of computer vision. Students will be assessed via a final project report, coupled with a demonstration and presentation.
The Robotics Institute is home to one of the largest academic groups of computer vision with expertise in relevant sub-ares, including sensing, computational photography, physics-based vision, tracking, 3D reconstruction, statistical analysis, object recognition, human modeling and analysis and general scene understanding. Students enrolled in the MSCV program will have access to world-class computer vision research facilities and a comprehensive list of courses offered by the faculty.

A Growing Field:

In recent years computer vision has changed the way we view the world. Some examples of computer vision applications include image-based Internet searches, street-view related applications, robotics, face recognition for social networks, safety systems on vehicles, visual product identification and searches, human-computer interfaces for visual communication and gaming, disease diagnostics using medical imaging, visual inspection of machine parts, visual crop quality assessment, etc. To support the rapid development of these applications, major companies including Adobe, Amazon, Apple, Canon, Facebook, GE, Google, IBM, Microsoft, NVIDIA, Qualcomm, Samsung and Siemens as well as numerous start-ups are forming computer vision groups.

MSCV Learning Outcomes:

Upon successful completion of the MSCV program students are expected to be proficient in:

Reading and understanding current research publications about state of the art computer vision techniques.
Using the fundamental development tools commonly used for developing computer vision applications.
Implementing computer vision applications based on state of the art algorithms.
Presenting the background and implementation details of a state of the art computer vision technique in a concise and clear manner.
Conducting experimental analysis and testing consistent with current practice in computer vision, including standard metrics and benchmark datasets.
Applying mathematical and machine learning tools, such as geometry, optimization, and statistics to computer vision applications.
For more information, contact ms-cv@ri.cmu.edu.

推荐课程

• 16-385  Computer Vision
• 16-423  Designing Computer Vision Apps
• 16-822 Geometry Based Methods in Computer Vision
• 16-623 Advanced Computer Vision Apps
• 16-823 Physics Based Methods in Vision
• 16-831 Statistical Techniques in Robotics