Image Processing and Image Synthesis

Project Overview

We are working on image processing and image synthesis. We have shown that deep networks can be used to quickly and accurately perform advanced image processing. We have developed deep network architectures that can synthesize photographic images, repeatedly advancing the realism of direct image synthesis. We have significantly advanced low-light imaging by applying deep networks directly to raw sensor data, in effect replacing much of the classic image processing pipeline; this technology has been widely adopted throughout the industry and is now in many smartphones. We are also introducing techniques for view synthesis with deep networks, allowing photorealistic exploration of complex large-scale scenes.


Enhancing Photorealism Enhancement

Dancing under the stars: video denoising in starlight

Stable View Synthesis

NeRF++: Analyzing and Improving Neural Radiance Fields

Free View Synthesis

Dynamic Low-light Imaging with Quanta Image Sensors

Seeing Motion in the Dark

Hiding Video in Audio via Reversible Generative Models

Acoustic Non-Line-of-Sight Imaging

Zoom to Learn, Learn to Zoom

Events-to-Video: Bringing Modern Computer Vision to Event Cameras

Semi-parametric Image Synthesis

Learning to See in the Dark

Interactive Image Segmentation with Latent Diversity

Photographic Image Synthesis with Cascaded Refinement Networks

Fast Image Processing with Fully-Convolutional Networks

Learning to Inpaint for Image Compression