YoVDO

SIREN - Implicit Neural Representations with Periodic Activation Functions

Offered By: Yannic Kilcher via YouTube

Tags

Machine Learning Courses Image Reconstruction Courses Hypernetworks Courses

Course Description

Overview

Dive into a comprehensive 56-minute video explanation of the paper "SIREN: Implicit Neural Representations with Periodic Activation Functions." Explore the concept of implicit neural representations and how SIRENs, a novel type of INR, can be applied to various signals like images, sound, and 3D shapes. Follow along as the video breaks down key concepts including initialization techniques, derivatives of SIRENs, Poisson image reconstruction and editing, and shapes with signed distance functions. Gain insights into other applications of SIRENs, hypernetworks, and their broader impact on the field. Benefit from a detailed outline covering topics from basic introductions to advanced concepts, making this explanation accessible for viewers with varying levels of expertise in machine learning and neural networks.

Syllabus

- Intro & Overview
- Implicit Neural Representations
- Representing Images
- SIRENs
- Initialization
- Derivatives of SIRENs
- Poisson Image Reconstruction
- Poisson Image Editing
- Shapes with Signed Distance Functions
- Paper Website
- Other Applications
- Hypernetworks over SIRENs
- Broader Impact


Taught by

Yannic Kilcher

Related Courses

Textual Inversion and Hypernetworks - Stable Diffusion 2
Nerdy Rodent via YouTube
Parameter Sharing - Recurrent and Convolutional Nets
Alfredo Canziani via YouTube
Stable Diffusion Style Technique Comparison - Hypernetwork vs. Textual Inversion
kasukanra via YouTube
Emergent Hypernetworks in Weakly Coupled Oscillators - IPAM at UCLA
Institute for Pure & Applied Mathematics (IPAM) via YouTube
Improving Pareto Front Learning via Multi-Head HyperNetwork
VinAI via YouTube