YoVDO

NeRF - Representing Scenes as Neural Radiance Fields for View Synthesis

Offered By: Yannic Kilcher via YouTube

Tags

Deep Learning Courses Neural Radiance Fields Courses Positional Encoding Courses

Course Description

Overview

Explore the groundbreaking Neural Radiance Fields (NeRF) technique for view synthesis in this comprehensive video explanation. Dive into the core concepts of NeRF, including its ability to embed entire scenes into neural network weights and achieve state-of-the-art results using sparse input views. Learn about the differential volume rendering procedure, directional dependence, and how NeRF captures fine structural details, reflections, and transparency. Follow along as the video breaks down the training process, radiance field volume rendering, positional encoding, and hierarchical volume sampling. Gain insights into the experimental results and understand how NeRF outperforms prior work in neural rendering and view synthesis.

Syllabus

- Intro & Overview
- View Synthesis Task Description
- The fundamental difference to classic Deep Learning
- NeRF Core Concept
- Training the NeRF from sparse views
- Radiance Field Volume Rendering
- Resulting View Dependence
- Positional Encoding
- Hierarchical Volume Sampling
- Experimental Results
- Comments & Conclusion


Taught by

Yannic Kilcher

Related Courses

Neural Networks for Machine Learning
University of Toronto via Coursera
機器學習技法 (Machine Learning Techniques)
National Taiwan University via Coursera
Machine Learning Capstone: An Intelligent Application with Deep Learning
University of Washington via Coursera
Прикладные задачи анализа данных
Moscow Institute of Physics and Technology via Coursera
Leading Ambitious Teaching and Learning
Microsoft via edX