DeepMind’s AI learned to turn flat images into 3D scenes

-
DeepMind’s AI learned to turn flat images into 3D scenes
By Tristan Greene


Google’s UK-based sister company, DeepMind, recently developed an AI that can render entire scenes in 3D after having only observed them as flat 2D images. Broad strokes: AI researchers at the cutting-edge are trying to teach machines to learn like humans. Rather than see the world in pixels, we look around our environment and make assumptions about everything in it. If we can see someone’s chest, we assume they also have a back, even though it may not be visible from our perspective. If you play peek-a-boo with a baby they learn that your face still exists even if you…

This story continues at The Next Web


June 29, 2018 at 09:59PM
via The Next Web https://ift.tt/2ICfHaW

Comments

Post a Comment

Popular posts from this blog

ASMR videos could be a new digital therapy for mental health

-
Image
ASMR videos could be a new digital therapy for mental health By The Conversation You may know “ASMR” as the niche genre of YouTube video which people watch on tablets and laptops to help them relax, perhaps before bed or in the lull of a Sunday afternoon. These videos typically involve someone role-playing a mundane professional service, such as giving you a haircut, a massage, or booking you in for a doctors’ appointment. The role-plays are usually performed in whispered voices, with the “ASMRtist” (the term for the creators of these videos) focused on building a feeling of intimacy and emphasizing any crisp sounds or slow hand movements. But what exactly is ASMR, and… This story continues at The Next Web October 4, 2018 at 01:08PM via The Next Web https://ift.tt/2DTwEkg
Read more

Fairies and deepfakes: a brief history of visual trickery

-
Image
Fairies and deepfakes: a brief history of visual trickery By The Conversation “Deepfake” is the name being given to videos created through artificially intelligent deep learning techniques. Also referred to as “face-swapping”, the process involves inputting a source video of a person into a computer, and then inputting multiple images and videos of another person. The neural network then learns the movements and expressions of the person in the source video in order to map the other’s image onto it to look as if they are carrying out the speech or act. This practice was first used extensively in the production of fake pornography in late 2017 – where the faces of… This story continues at The Next Web September 25, 2018 at 07:13PM via The Next Web https://ift.tt/2Oc2SLs
Read more

This Hacker News thread is a masterclass in how to put down your damn smartphone

-
Image
This Hacker News thread is a masterclass in how to put down your damn smartphone By Bryan Clark While it’s not worth rehashing the details, I believe most of us would agree that smartphones are both good and bad. On the one hand, we have a device that holds the vast majority of human knowledge. It’s an invention responsible for a shift in power that democratizes information and empowers those most capable of dispensing it — in theory, at least. On the other, we have a doomsday device responsible for wrecking societal norms (for better or worse) and leading to a population that’s sleep deprived, anxious, and generally unhappy — among other things. Ignorance, as they say, is… This story continues at The Next Web April 25, 2018 at 04:31AM via The Next Web https://ift.tt/2HWveUq
Read more