Inventor

AMIRGHODSI SOHRAB

US55 patents

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

AMIRGHODSI SOHRAB

ADOBE INC

Texture interpolation using neural networks

Labeling techniques for a modified panoptic labeling neural network

Generating modified digital images using deep visual guided patch match models for image inpainting

Generating modified digital images via image inpainting using multi-guided patch match and intelligent curation

Object class inpainting in digital images utilizing class-specific inpainting neural networks

Learning parameters for generative inpainting neural networks utilizing object-aware training and masked regularization

Digital image inpainting utilizing plane panoptic segmentation and plane grouping

Generating deep harmonized digital images

Generating modified digital images utilizing nearest neighbor fields from patch matching operations of alternate digital images

Controlling smoothness of a transition between images

Content aware sampling during patch synthesis

Controlling smoothness of a transition between images

Labeling techniques for a modified panoptic labeling neural network

Image inpainting based on multiple image transformations

Automatic synthesis of a content-aware sampling region for a content-aware fill

Generating iterative inpainting digital images via neural network based perceptual artifact segmentations

Learning parameters for an image harmonization neural network to generate deep harmonized digital images

Generative image filling using a reference image

Mask-robust image inpainting utilizing machine-learning models

Generating neural network based perceptual artifact segmentations in modified portions of a digital image

Improving digital image inpainting utilizing plane panoptic segmentation and plane grouping

Generating modified digital images utilizing nearest neighbor fields from patch matching operations of alternate digital images

Learning parameters for neural networks using a semantic discriminator and an object-level discriminator

Iteratively modifying inpainted digital images based on changes to panoptic segmentation maps

Image inpainting with geometric and photometric transformations

Digital image inpainting utilizing a cascaded modulation inpainting neural network

Performing patch matching guided by a transformation gaussian mixture model

Image inpainting with geometric and photometric transformations

Generating modified digital images by identifying digital image patch matches utilizing a Gaussian mixture model

Patch validity test

Image and object inpainting with diffusion models

Image inpainting based on multiple image transformations

Automatic synthesis of a content-aware sampling region for a content-aware fill

Image reprojection and multi-image inpainting based on geometric depth parameters

Generating deterministic digital image matching patches utilizing a parallel wavefront search approach and hashed random number

Previewing a content-aware fill

Panoptically guided inpainting utilizing a panoptic inpainting neural network

ADOBE SYSTEMS INC

Compact color feature vector representation

Compact color feature vector representation

Image resampling using variable quantization bins

Compact color feature vector representation

Content aware sampling during patch synthesis

Controlling smoothness of a transmission between images

Image synthesis utilizing an active mask

Cloud based content aware fill for images

Seamless compositing using a soft selection

Seamless compositing using a soft selection

Low memory content aware image modification

Image resampling with geometric transformation

AMIRGHODSI SOHRAB

Image resampling with geometric transformation