校際選修

115-1 選課時程

進行中

  • 初選第一階段 6/15/2026
  • 初選第二階段 6/22/2026
  • 校際選修 8/24/2026
  • 初選第三階段 8/31/2026
  • 開學後加退選 9/7/2026
  • 逾期加退選 9/21/2026
選課資源

基於深度學習之視覺辨識專論

Selected Topics in Visual Recognition using Deep Learning

學期
114-2
學分
3 學分
當期課號
535521
永久課號
CSIC30035
開課單位
資訊科學與工程研究所
授課教師
林彥宇
校區
光復
類別
選修
上課時間表
週二
5
13:20–14:10
基於深度學習之視覺辨識專論
EC114
3 節連堂
6
14:20–15:10
7
15:30–16:20

* 根據陽明交大上課時間表所列

概述

Computer vision aims to empower computers with the ability to "see" – to perceive, understand, and interpret the visual world much like humans do. Deep learning has emerged as the driving force behind the current computer vision revolution. The availability of massive, annotated datasets, coupled with the accessibility of powerful GPUs, has enabled the training of complex deep learning models. These models, consisting of hundreds of layers and millions of parameters, have significantly advanced the performance of numerous computer vision applications. In this course, we will begin by exploring key deep learning architectures crucial for computer vision research. This will include a deep dive into foundational concepts such as deep neural networks, convolutional neural networks (CNNs), Transformers, and denoising diffusion models. Subsequently, we will delve into several important computer vision applications, including object recognition, detection, segmentation, low-level vision, and 3D vision. For each application, we will examine the state-of-the-art deep learning algorithms that are driving progress in that area.

先修科目

1. Foundational mathematical skills, including linear algebra and calculus 2. Programming experience with Python and common libraries 3. Deep learning programming skills with frameworks like PyTorch

評分方式

Four homework assignments 64% (=16% x 4) Final project 36%

週次計畫
週次主題
第 1 週Introduction
第 2 週Deep Neural Networks
第 3 週Convolutional Neural Networks
第 4 週Transformers
第 5 週Object Detection I
第 6 週Object Detection II
第 7 週Object Segmentation I
第 8 週Object Segmentation II
第 9 週Denoising Diffusion Models
第 10 週Low-level Vision
第 11 週Mamba
第 12 週3D Point Clouds, Neural Radiance Fields (NeRF), and 3D Gaussian Splatting (3DGS)
第 13 週3D Vision
第 14 週Guest Lecture (Date subject to change based on speakers' schedule)
第 15 週Final Project Presentation I
第 16 週Final Project Presentation II
教科書

1. Ian Goodfellow, Yoshua Bengio, and Aaron Courville, Deep Learning, MIT Press, 2016 2. Richard Szeliski, Computer Vision: Algorithms and Applications, Springer, 2022

Office Hours
地點
EC706 (Instructor) EC234-C or EC701 (TAs) Please send us an email in advance to make an appointment, and we will inform you where to have a discussion.
時間
Tuesday 4:20 pm ~ 5:20 pm
聯絡方式
Instructor: Yen-Yu Lin (林彥宇) Email: lin@cs.nycu.edu.tw TAs: Tsung-Lin Tsai (蔡宗霖) Email: sean19990323123@gmail.com Jian-Zhe Wang (王健哲) Email: jzwang.cs13@nycu.edu.tw Yi-Jen Tsai (蔡宜蓁) Email: tsai.cs14@nycu.edu.tw Nai-Yun Hsiao (蕭乃云) Email: alllllvin21292@gmail.com