This workshop teaches deep learning techniques to design, train, and deploy neural networks for digital content creation through a series of hands-on exercises. You will work with widely-used deep learning tools, frameworks, and workflows by performing neural network training on a fully-configured, GPU-accelerated workstation in the cloud. The workshop teaches techniques for transferring the look and feel of an image, as well as autoencoder- based techniques to enhance image quality.
You’ll also learn methods to detect noise and train your own denoiser on some sample images. By the end of the workshop, you will be able to create paintings from your photographs, make noise-free images from noisy ones, and convert low-resolution images to high-resolution images.
At the conclusion of the workshop, you will have an understanding of autoencoders and be able to:
- Explore the architectural innovations and training techniques used to make arbitrary photo and video style transfer.
- Train your own denoiser for rendered images.
- Train a network to create high-resolution image from low-resolution ones.
Why Deep Learning Institute Hands-On Training?
- Learn how to build deep learning and accelerated computing applications across a wide range of industry segments such as autonomous vehicles, digital content creation, finance, game development, and healthcare
- Obtain guided hands-on experience using the most widely-used, industry-standard software, tools, and frameworks
- Gain real-world expertise through content designed in collaboration with industry leaders including the Children’s Hospital Los Angeles, Mayo Clinic, and PwC
- Earn NVIDIA DLI Certification to demonstrate your subject matter competency and support professional career growth
- Access content anywhere, anytime with a fully-configured, GPU-accelerated workstation in the cloud
- Upon successful completion of the workshop, participants will receive NVIDIA DLI Certification to recognize subject matter competency and support professional career growth
- Getting started
Instructor introduction and environment setup
Image Style Transfer with Torch
- Transfer the look and feel of one image to another image by extracting distinct visual features
- Qualitatively determine whether a style is transferred correctly using different techniques
- Use architectural innovations and training techniques for arbitrary style transfer
Explore the architectural innovations and training techniques used to make arbitrary photo and video style transfer
Image Super-Resolution Using Autoencoders
- Understand and design an autoencoder
- Train and run a model to produce high-quality images from low-quality ones
- Learn various methods to rigorously measure image quality
Train your own denoiser for rendered images
Rendered Image Denoising Using Autoencoders
- Determine whether noise exists in rendered images
- Use a pre-trained network to denoise sample images, or your own images
- Train your own denoiser using the provided dataset
Train a network to create high-resolution images from low-resolution ones
- Answer multiple choice questions covering all three labs
Test your understanding of the material and identify any gaps in knowledge
- Tools, libraries, and frameworks: TensorFlow, Torch
- Basic familiarity with deep learning concepts, such as convolutional neural networks (CNNs)
- Experience programming in Python