電腦視覺

This course aims to provide students with an in-depth understanding of the fundamental concepts and techniques in computer vision, including image formation, image feature extraction, 3D reconstruction, image segmentation, object recognition, deep learning, object detection, object tracking, and face recognition. Students will also gain proficiency in implementing and applying algorithms, models, and frameworks related to computer vision, equipping them to solve computer vision problems in various fields such as autonomous driving, smart homes, and medical image analysis. In addition, the course features a special joint lecture by Dr. Chia-Ming Cheng (鄭嘉珉) from MediaTek’s AI Technology Division, who will share the latest industry insights and perspectives, offering students an in-depth exploration of real-world challenges and business opportunities in both research and end-product applications of computer vision. Through this course, students will understand the limitations and challenges of existing computer vision applications and explore future development directions. By the end of the course, students will have developed the following abilities:  Students will possess basic knowledge and technical skills in computer vision and image processing.  Students will have sensitivity to emerging technologies and trends, along with strong analytical abilities.  Students will be capable of conducting independent research and development work, with competencies in teamwork and project management.  Students will have innovative thinking and problem-solving abilities, applying what they have learned to promote technological innovation and social progress.

Deep Learning, Python

 Lectures on Theories and Principles: Instructional sessions covering the fundamental theories and principles of computer vision.  Practical Sessions and Case Analyses: Hands-on activities and analysis of real-world cases to apply theoretical knowledge.  Student Presentations and Discussions: Opportunities for students to present their work and engage in discussions to enhance understanding and critical thinking.

 Assignments (40%): Including programming assignments, literature review reports, etc.  Midterm Report (20%): Students are required to select a computer vision-related paper from the past three years and write a research report. Additional points will be awarded if the report includes a demo and technical implementation.  Final Project (40%): Students must choose a computer vision-related topic and produce both a research report and a practical implementation project. The grading criteria include a clear understanding of the problem, the innovation and practicality of the solution, and the completeness and effectiveness of the technical implementation.  Attendance/Class Participation: -1 each absent.

Programming Computer Vision with Python, Jan Erik Solem, O’REILLY Media, June 2012. (ISBN: 9781449316549)

地點

374, Engineering Building-VI

時間

11:00~12:00, Monday

聯絡方式

yfyangd@gmail.com