Inventor

PERAZZI FEDERICO

US23 patents

⚠️ This page may combine multiple inventors who share the name “PERAZZI FEDERICO”. Patents are grouped by organization below to help distinguish them — per-person disambiguation is on the roadmap.

PERAZZI FEDERICO

ADOBE INC

Utilizing a two-stream encoder neural network to generate composite digital images

Utilizing a neural network having a two-stream encoder architecture to generate composite digital images

Robust training of large-scale object detectors with a noisy dataset

Neural network architecture pruning

Retouching digital images utilizing separate deep-learning neural networks

Temporally distributed neural networks for video semantic segmentation

Denoising images rendered using Monte Carlo renderings

Retouching digital images utilizing layer specific deep-learning neural networks

Temporally distributed neural networks for video semantic segmentation

Generating and utilizing pruned neural networks

Kernel prediction with kernel dictionary in image denoising

Kernel prediction with kernel dictionary in image denoising

Determining video cuts in video clips

Shaping a neural network architecture utilizing learnable sampling layers

DISNEY ENTPR INC

Methods and systems of performing video object segmentation

Video super-resolution using an artificial neural network

Visual saliency estimation for images and video

Image processing using a convolutional neural network

Panoramic video from unstructured camera arrays with globally consistent parallax removal

Video segmentation from an uncalibrated camera array

Systems and methods for higher order dimensional space video segmentation

PERAZZI FEDERICO

Visual saliency estimation for images and video

BEARDSLEY PAUL

Robotic texture