YoVDO

Human Activity Recognition - Learning with Less Labels and Privacy Preservation

Offered By: University of Central Florida via YouTube

Tags

Machine Learning Courses Data Labeling Courses Contrastive Learning Courses

Course Description

Overview

Explore cutting-edge approaches to Human Activity Recognition in this keynote talk from SPIE Automatic Target Recognition XXXII. Delve into innovative techniques for learning with limited labeled data and preserving privacy. Examine problem statements, results, and various approaches including privacy leakage prevention, pseudoleveling, and contrastive learning. Discover the potential of Zero Shot Learning and Local Local Contrast Loss in this field. Analyze experimental and qualitative results from cell surveys, gaining valuable insights into the future of activity recognition technology.

Syllabus

Intro
Problem Statement
Results
Problem
Approaches
Privacy leakage
Pseudoleveling
Contrasting
Local Local Contrast Loss
Zero Shot Learning
Evaluation
Cell Survey
Experimental Results
Qualitative Results


Taught by

UCF CRCV

Tags

Related Courses

Stanford Seminar - Audio Research: Transformers for Applications in Audio, Speech and Music
Stanford University via YouTube
How to Represent Part-Whole Hierarchies in a Neural Network - Geoff Hinton's Paper Explained
Yannic Kilcher via YouTube
OpenAI CLIP - Connecting Text and Images - Paper Explained
Aleksa Gordić - The AI Epiphany via YouTube
Learning Compact Representation with Less Labeled Data from Sensors
tinyML via YouTube
Robust Pre-Training by Adversarial Contrastive Learning - CAP6412 Spring 2021
University of Central Florida via YouTube