Category: AI

Become a MetaHuman

 

 

Epic’s MetaHuman Avatar tool now has the abilty to import scans of real people, this is open for developers currently, but watch this space as I will predict it will not be long before you are able to create your own metahuman that looks like your twin.

 

 

To caveat this was found via Roadtovr.com and they quoted this

“There’s still some limitations, however. For one, hair, skin textures, and other details are not automatically generated; at this point the Mesh to MetaHuman feature is primarily focused on matching the overall topology of the head and segmenting it for realistic animations. Developers will still need to supply skin textures and do some additional work to match hair, facial hair, and eyes to the person they want to emulate.

The MetaHuman tool is still in early access and intended for developers of Unreal Engine.”

Advancement in Digital Humans

Lastly, this week I found this company Pantheonlab.ai that are creating digital humans. If you have read my other articles here:

Digital Humans taking over

Digtial Humans taking over Cont..

You will see I am a little obsesses with digital humans and the value they can add with production and services like help desks. 

Pantheon Lab have created very realistic digital humans that can also switch gender, appearance, or race.

With the advancement of technology, it is now possible to create artificial characters that just look real, shift faces, sync lips, or clone voices.

And NO this person isn’t real in the video..

 

 

Live Portraits

We have all seen deepfakes and how real they can seem, a group have taken it a step further and the resolution and small details like month and eye movement are really impressive.

A snippet from the website

“We propose a system for the one-shot creation of high-resolution human avatars, called megapixel portraits or MegaPortraits for short. Our model is trained in two stages. Optionally, we propose an additional distillation stage for faster inference.

Our training setup is relatively standard. We sample two random frames from our dataset at each step: the source frame and the driver frame. Our model imposes the motion of the driving frame (i.e., the head pose and the facial expression) onto the appearance of the source frame to produce an output image. The main learning signal is obtained from the training episodes where the source and the driver frames come from the same video, and hence our model’s prediction is trained to match the driver frame. “

Website: https://samsunglabs.github.io/MegaPortraits/

 

 

 

Chat with Einstein the digital human

Lately I have taken a massive interest into Digital Humans and all they can offer using AI, machine learning and buckets of data.
This example is spot on using a figure that the world know so well and bringing him back to life for classrooms to ask him questions on all of his amazing work and his life. He responds in a casual way that doesn’t feel too forced enabling you to actually start a conversation, i would recommend using your mic vs typing to give the conversations a natural feel.

Overall I was very impressed when i have my little chat with one of my heros.

Audio content production company Aflorithmic and digital humans company UneeQ teamed up to create a digital version of the famous genius, Albert Einstein. – See the video below and to chat to him click this LINK

Source: Interesting Engineering

Five AI developments will shape 2021 and beyond

1. AI and vaccine development
It typically takes years, if not decades, to develop a new vaccine. But by March 2020, vaccine candidates to fight covid-19 were already undergoing human tests, just three months after the first reported cases.

2. Fully automated driving and the rollout of robotaxis
Autonomous driving technology continued to mature in 2020, with the industry’s leading companies testing driverless cars and opening up robotaxi services to the public in various cities.

3. Applied natural language processing
In 2020, natural language systems became significantly more advanced at processing aspects of human language like sentiment and intent, generating language that aligns with human speaking and writing patterns, and even visual understanding, meaning the capability to express understanding about an image through language.

4. Quantum computing
Quantum computing has the potential to supercharge AI applications compared to binary-based classical computers

5. AI chips
Chips can already cope with a small amount of AI processing, but with the need for more AI we will need to develop chips that can focus on the deep learning processing required.

Read the full article on MIT