Capturing Deep-Learning Data for Neural Network Training

Channel:
NVIDIA Game Developer
Subscribers:
20,000
Published on ● Video Link: https://www.youtube.com/watch?v=mDlqLfRGmYk


Duration: 8:10
407 views
6

At GDC 2019, NVIDIA presented a talk called “Truly Next Gen: Adding Deep Learning to Games & Graphics”, which detailed the current state of the art in deep learning for games. In this excerpt, Anjul Patney, Senior Research Scientist at NVIDIA, provides information on how to capture data for neural network training. He also describes common deep learning bugs, and explains how to fix them.



Other Videos By NVIDIA Game Developer


2019-08-07Making Realistic Waves with WaveWorks 2
2019-08-01NVIDIA Real Time Rendering of “Measure 1” from the Zero Day Short Film
2019-07-29NVIDIA Studio Sessions | Enable RTX Ray Tracing In Unreal Engine 4
2019-07-29Introducing: Nsight Aftermath SDK
2019-07-29Nsight Graphics 2019.4
2019-07-18Material Workflows in Omniverse with NVIDIA MDL
2019-07-18Mesh Shading with NVIDIA at SIGGRAPH 2019
2019-07-16Ray Tracing with NVIDIA at SIGGRAPH 2019
2019-05-28Nsight Graphics 2019.3
2019-04-25Ray Tracing's Rapid Evolution
2019-04-25Capturing Deep-Learning Data for Neural Network Training
2019-04-25An Engineer's Guide to Integrating Ray Traced Irradiance Fields
2019-04-23Nsight Systems - Vulkan Trace
2019-03-25NVIDIA Announced Nsight Graphics 2019.2
2019-03-19WaveWorks 2.0: Three Things you Need to Know
2019-03-19Ray Tracing for Vulkan: Three Things you Need to Know
2019-03-19DXR Spotlight Contest: Three Things you Need to Know
2019-03-19Grapeshot Games Explain Why They Chose WaveWorks 2.0 for Atlas
2019-03-18DLSS: Three Things you Need to Know
2019-03-14Nsight Graphics: Three Things You Need to Know
2019-03-13Mesh Shading: Three Things You Need to Know


Tags:
GDC
NVIDIA
DLSS
Deep Learning
Games
Graphics
Neural Network Training
Gaming
PC
PC Gaming