深度學習應用於電腦視覺 Deep Learning in Computer Vision
Course Introduction
Deep Learning in Computer Vision
114-2 Elective course (3.0 credits). Machine learning, deep learning, and computer vision algorithms with engineering applications.
✦ Course Information
| Course title | Deep Learning in Computer Vision |
|---|---|
| Semester | 114-2 |
| Designated for | COLLEGE OF ENGINEERING DEPARTMENT OF CIVIL ENGINEERING |
| Curriculum Number | CIE5151 |
| Curriculum Identity Number | 521EU9310 |
| Class | — |
| Credits | 3.0 |
| Full / Half Yr. | Half |
| Required / Elective | Elective |
| Remarks | Restriction: within this department (including students taking minor and dual degree program) The upper limit of the number of students: 50. |
| Table of Core Capabilities and Curriculum Planning | Association has not been established |
Please respect the intellectual property rights of others and do not copy any of the course information without permission.
Class Section
| Class | Instructor | Time | Student Workload |
|---|---|---|---|
| — | RIH-TENG WU | Monday 6 (13:20–14:10) Thursday 5, 6 (12:20–14:10) |
4hrs |
Course Description
This course introduces the fundamental theory/background knowledge of prevalent machine learning (ML) and computer vision (CV) algorithms. Relevant applications in the broad domain of the engineering community will be introduced to motivate the students.
The first half of the semester will focus on the reasoning of artificial intelligence, several ML algorithms, model evaluation, deep learning (DL) and reinforcement learning. The rest of the semester will have emphasis on the reasoning of image processing, image feature extractions and pairing, as well as image-based sensing.
After taking this course, students are expected to be equipped with basic knowledge and implementation skills to develop ML, DL or CV based approaches for applications in engineering.
Course Objective
Upon taking this course, students are anticipated to be well-prepared in the following items:
Understand the fundamental principles that support the ML/DL algorithms.
Be able to reasoning the performance of ML/DL models.
Be able to implement ML/DL algorithms.
Understand the fundamental principles that support the CV algorithms.
Understand the image representations of the world.
Be able to implement CV algorithms.
Course Requirement
- Course Requirement: Prerequisites: Calculus, Computer Programming
- Student Workload (Expected weekly study hours before and/or after class): 4hrs
- Office Hours: Thu. 14:30~16:30 Note: Absence of the class will be allowed only if the student informed the instructor in advance. Contact(TA) : r13521629@ntu.edu.tw
- Designated reading: —
References
Several excellent online sources are:
A Course in Machine Learning, electronic source available at: http://ciml.info/
Christopher Bishop (2006), Pattern Recognition and Machine Learning, Springer
Goodfellow et. al (2016), Deep Learning, MIT Press, electronic source available at: https://www.deeplearningbook.org/
Grading
| No. | Item | % | Explanations for the conditions |
|---|---|---|---|
| 1. | Term project | 30% | |
| 2. | Assignment | 35% | |
| 3. | Midterm | 30% | |
| 4. | Participation | 5% |
- Teaching methods: Assisted by recording, Assisted by video, Provide students with flexible ways of attending courses
- Assignment submission methods: Mutual agreement to present in other ways between students and instructors
- Exam methods
- Others
Grading Policy
NTU has not set an upper limit on the percentage of A+ grades.
Letter Grade System
NTU uses a letter grade system for assessment. The grade percentage ranges and the single-subject grade conversion table in the NATIONAL TAIWAN UNIVERSITY Regulations Governing Academic Grading are for reference only.
Instructor Adjustment
Instructors may adjust the percentage ranges according to the grade definitions. For more information, see the Assessment for Learning Section.
Progress
| Week | Date | Topic |
|---|---|---|
| Week 1 | 2/23, 2/26 | Introduction to artificial intelligence, machine learning, and deep learning |
| Week 2 | 3/2, 3/5 | Data representations; Evaluation of machine learning models |
| Week 3 | 3/9, 3/12 | Support vector machine |
| Week 4 | 3/16, 3/19 | Support vector machine (Cont.); k-nearest neighbor |
| Week 5 | 3/23, 3/26 | Decision tree; Fully-connected neural network |
| Week 6 | 3/30, 4/2 | Fully-connected neural network (Cont.) |
| Week 7 | 4/6, 4/9 | 4/6 (break); Introduction to image basics, image-based sensing, image filtering |
| Week 8 | 4/13, 4/16 | Image filtering (Cont.); Convolutional neural network |
| Week 9 | 4/20, 4/23 | Convolutional neural network (Cont.) |
| Week 10 | 4/27, 4/30 | Transfer learning; Auto-encoder |
| Week 11 | 5/4, 5/7 | Generative adversarial network; Midterm (5/1) |
| Week 12 | 5/11, 5/14 | Object classification, detection and segmentation |
| Week 13 | 5/18, 5/21 | Feature extraction and pairing |
| Week 14 | 5/25, 5/28 | Digital image correlation and image stitching |
| Week 15 | 6/1, 6/4 | World-image correspondence |
| Week 16 | 6/8, 6/11 | 3D reconstruction (optional) |
