Skip to main content

LH Computer Vision and Imaging - Lab 5

info

The lab sheet and files can be downloaded here

Introduction 介紹​

In this lab exercise, you will look at the idea behind face recognition using a dimensionality reduction method called PCA. You are asked to write a short (no more than 2 pages) report of your work, answering specific questions, and showing example images. This work is not assessed (it will not count towards your module mark) but you will get formative feedback. 在本實驗練習中，您將使用稱為 PCA 的降維方法了解人臉識別背後的理念。你被要求寫一份簡短的（不超過 2 頁）工作報告，回答具體問題，並展示示例圖片。這項工作不會被評估（它不會計入你的模塊分數）但你會得到形成性的反饋。

STEP 1​

  • Download the zip file and extract the data files for Lab from CANVAS and save them in your working directory 從 CANVAS 下載 zip 文件並提取 Lab 的數據文件，並將它們保存在您的工作目錄中

TASK​

  • Work through the matlab script ‘eigen_faces.m’ that has comments to explain all the steps. 完成 matlab 腳本“eigen_faces.m”，其中包含解釋所有步驟的註釋。

Question 1​

  • One line 38 when you plot a face, why is it inverted? Does it matter? 一行 38 畫人臉的時候，為什麼是倒著的？有關係嗎？

Question 2​

  • When we need to calculate PCA for higher resolution we need to think differently on how to improve computational speed! How can we achieve this? 當我們需要計算更高分辨率的 PCA 時，我們需要以不同的方式思考如何提高計算速度！我們怎樣才能做到這一點？

Question 3​

  • When you look at each PC, in order of importance, what do you see? 當你看到每個 PC 時，按重要性排序，你看到什麼？

Question 4​

  • As you decrease the number of PCs used for face reconstruction, what do you observe? 當你減少用於人臉重建的 PC 數量時，你觀察到什麼？

Question 5​

  • Based on the lecture notes, modify the algorithm to allow face recognition. 根據講義，修改算法以允許人臉識別。