Computer Vision: Classification and Object Recognition - Lecture 11

Explore the fundamentals of image classification in computer vision through this comprehensive 47-minute lecture from the University of Central Florida's CAP5415 course. Delve into object recognition techniques and the ImageNet dataset, understanding the importance of dataset splitting and feature extraction. Learn about the machine learning framework for classification tasks, focusing on nearest neighbor and linear classifiers. Examine decision boundaries, K-nearest neighbor algorithms, and the motivation behind linear classifiers. Gain insights into classifier design principles, including maximum margin concepts and Support Vector Machines (SVM). This lecture provides a solid foundation for understanding classification techniques in computer vision applications.

Intro
Classification - computer vision
Object Recognition
Image classification - ImageNet
Dataset split
Features
The machine learning framework
Classifiers: Nearest neighbor
Decision boundary for NN Classifier
K-nearest neighbor
Algorithm
Classifiers: Linear
Motivation
Classifier Design
Maximum Margin
SVM - background
Maximal-Margin Classifier