Articles tagged "artificial-intelligence"

A team of researchers at Waymo have created Block-NeRF, a new AI technique that uses neural radiance fields to create large scalable scene neural view synthesis. Block-NeRF has been used to recreate an entire San Francisco neighborhood.

This novel approach integrates its neural radiance field (NeRF) into a style-based generator to improve rendering efficiency, resolution and 3D consistency compared to previous models.

In 2019, NVIDIA showed off GauGAN, an AI-powered demo that generates images from sketches. NVIDIA has been training GauGAN since then, and the AI can now respond to semantic prompts.

Skylum has revealed the next version of its Luminar photo-editing software, Luminar Neo. The new app includes multiple AI-based editing tools, improved performance and a new editing engine.

ImagenAI recently announced its new desktop app for Mac and PC. Using artificial intelligence, it will automatically edit batches of your photos in Lightroom based on what it has learned from your past editing preferences.

Genealogy company MyHeritage has launched a new AI-powered service, Deep Nostalgia. The AI tech, licensed from D-ID, analyzes the faces in photographs and uses deep learning to apply animation sequences.

Canon is bringing one of its recent concept cameras into reality. The PowerShot PICK is an AI-powered camera capable of automatically tracking and photographing subjects thanks to its 360º pan/tilt design.

Building upon its GPT-3 language model, OpenAI's new AI creation, DALL-E, can create images of anything you can describe with varying levels of success. If you want an image of a turtle in a forest from the perspective of a fisheye lens, DALL-E can do it.

BlinkAI's patented AI-powered image enhancement technology works alongside the Qualcomm Snapdragon 888 SoC inside the Mi 11 smartphone to get the most out of low-light video footage.

Google has asked its users for help improving Google Photos before, but a new participation option for Android users pushes this idea even further. You can now volunteer to label a group of images to help improve Google's machine learning algorithms.

Kodak Professional Select is a new image selection software. It relies upon AI to analyze, grade and select the best images from a given batch.

Skylum Software has teased a new feature for its upcoming photo editor, Luminar AI. In 2021, Luminar AI's AI Sky feature will be able to create realistic reflections in water. This feature has been requested by users since Luminar's first AI Sky Replacement tool launched in 2019.

The AI is already available to Azure customers and will be incorporated into multiple Microsoft consumer products later this year.

The images won't be mistaken for actual photos, but they do reveal the machine's understanding of the visual world and how it relates to concepts presented as text captions.

The tool was announced as part of Microsoft's effort to help prevent election interference in the form of digital manipulation.

Skylum Software has unveiled the newest version of Luminar. LuminarAI has been designed around the idea of using artificial intelligence to simplify as much of the photo editing process as possible without removing the creative control of the user.

The technology involves two machine learning models, one to remove unwanted shadows and another to soften natural facial shadows.

The newly detailed work adds extensions to an available AI technology, giving it the new ability to transform tourist images into complex, accurate models of the subject.

The utility is intended to poison existing facial recognition models by imperceptibly altering pixels in an image of a person's face, disrupting a machine's ability to recognize them in non-cloaked images.

Researchers with IBM and MIT have published a new demonstration tool that allows anyone to 'paint' new elements into existing digital images.

The company says it looking into providing ‘new imaging experiences’ with technologies such as artificial intelligence.

GauGAN as it's known, can create photorealistic images from basic drawings using the power of artificial intelligence.

Researchers have developed a new web app, powered by artificial intelligence, that colorizes photos with incredible accuracy.

NVIDIA researchers have created an AI that can create realistic portraits on demand using various elements from source images.

Xiaomi has laid out the details for its new AI-powered image processing platform DeepExposure.

A groups of researchers from NVIDIA, MIT, and Aalto University have developed an AI capable of removing noise and grain from images with incredible accuracy.

Researchers at the University of Illinois Urbana–Champaign and Intel have developed a neural network that can brighten extremely low-light images without adding any noise. It's not an exaggeration to say the results are incredible.

The team at BBC's Tomorrow's World built a Facebook Messenger chatbot that uses artificial intelligence to analyze and score your photos based on color, composition, emotion, landmarks, and more.

Google relied on the expertise of a documentary filmmaker, a photojournalist, and a fine arts photographer to train the AI and feed some high-quality photography into its machine learning model.

An AI-based system developed at the Max Planck Institute for Intelligent Systems is capable of creating high-definition versions of low-resolution images, using a texture-based approach.