🚀 Emerging from the Uncanny Valley: Time to Sharpen Our Digital Discernment 🚀 With AI and digital media evolving quickly these days, Microsoft's latest project (https://lnkd.in/gzYATkx4) presents us with hyper-realistic videos created from just an image and a script, challenging our perceptions of what's real and what's computer-generated. It seems that we might be about to exit our 2 decade journey through Masahiro Mori's Uncanny Valley. This clearly calls for increase awareness and skill in spotting deep fakes. Here are things I added to my mental list of digital tells in this new era of everyday digital forensics🔍: Gravity-Defying Hair: Notice how long hair in these videos often seems to ignore the laws of physics? It's a subtle, yet telltale sign of digital artifice. Eerie Ears: Watch for ears that seem to oddly float or morph, especially during movement. This is a common glitch in the realm of CGI. Facial Musculature: A lack of natural movement in facial muscles, particularly during expressive dialogue, can be a dead giveaway. #DigitalMedia #TechInnovation #AI #Microsoft #VisualEffects #ProfessionalDevelopment #UncannyValley
Johan Bouvin’s Post
More Relevant Posts
-
Imagine being able to generate a hyper-realistic talking face video from just a single portrait photo and a speech audio clip 🤯 With VASA-1, you can do just that! VASA-1 created by Microsoft produces lifelike talking faces with precise lip-audio sync, natural facial behavior, and realistic head movements, all in real-time ⏱️. You can even control the generation with optional signals like eye gaze direction, head distance, and emotion offsets 🎨. The possibilities are endless 🚀 From enhancing educational equity to offering companionship, this tech has the potential to make a real impact 🌎. Check out the page and see for yourself 👉 https://lnkd.in/gGyA9Uek #VASA-1 #TalkingFaces #AI #ResponsibleAI #GenerativeAI #ArtificialIntelligence
VASA-1 - Microsoft Research
https://www.microsoft.com/en-us/research
-
This is by far one of the best demos I have seen on AI by Microsoft Research Asia Team in the recent times - https://lnkd.in/d2kS4XMe VASA-1: Lifelike Audio-Driven Talking Faces Generated in Real Time TL;DR: single portrait photo + speech audio = hyper-realistic talking face video with precise lip-audio sync, lifelike facial behavior, and naturalistic head movements, generated in real time. Please check out the videos. Enjoy!! #Microsoft #AI #Research
VASA-1 - Microsoft Research
https://www.microsoft.com/en-us/research
-
VASA-1 is a Microsoft research project that generates lifelike talking faces in real time, using only a single static image and a speech audio clip. Whilst it is fun to see Mona Lisa rapping, and exciting to see another AI advancement, this is also a reminder of the needs for responsible development and use. This research is intended to focus on visual affective skills for virtual AI avatars, aiming for positive applications. It is not intended to create content that is used to mislead or deceive. As of now, there is no plan for commercial products or API or online demo. https://lnkd.in/eheMU7By #MicrosoftAI
VASA-1 - Microsoft Research
https://www.microsoft.com/en-us/research
-
Microsoft has released VASA-1: A tool based on diffusion model that takes a single photo and speech audio and generates a realistic talking face video with precise lip-audio sync, lifelike facial behavior, and naturalistic head movements, generated in real time. The model accepts optional signals as condition, such as main eye gaze direction, head distance, and emotion offsets Example: https://lnkd.in/dpbHibmK More. examples: https://lnkd.in/dzQTD9fp Paper: https://lnkd.in/d_9Bswvh #vision #diffusion #model #ai #lipsync #realtime #machinelearning #ml
VASA-1 - Microsoft Research
https://www.microsoft.com/en-us/research
-
🚀 Exciting Update from Microsoft Research! 🚀 Check out VASA-1, a breakthrough framework by Microsoft that brings static images and speech clips to life as lifelike talking faces. VASA-1 excels in lip-sync, facial nuances, and head motions, offering real-time video generation at high resolutions and speeds. Learn more about how this innovation is setting new standards in virtual interactions: https://lnkd.in/d6C_6RMg #MicrosoftResearch #AI #VirtualReality #Innovation
VASA-1 - Microsoft Research
https://www.microsoft.com/en-us/research
-
This is amazing research by Microsoft on generating realistic talking faces from a single image and speech audio link to the article. Imagine creating virtual assistants that can not only respond to your queries but also deliver them with natural facial expressions and head movements! This technology has the potential to revolutionize the way we interact with machines. However, it's important to acknowledge the potential downsides of this technology, as it could be misused to create deepfakes. We need to be mindful of these potential risks as this technology develops. #microsoft #ai #deeplearning #machinelearning #virtualassistants https://lnkd.in/gcmSfnYe
VASA-1 - Microsoft Research
https://www.microsoft.com/en-us/research
-
Another day, another AI model that is groundbreaking. To quote the Microsoft paper: "We introduce VASA, a framework for generating lifelike talking faces of virtual characters with appealing visual affective skills (VAS), given a single static image and a speech audio clip." Have a look at the videos on the link below: https://lnkd.in/eTRjgdN8 #ai
VASA-1 - Microsoft Research
https://www.microsoft.com/en-us/research
-
Crazy tech, MS won't release it yet, concerned about deep fake issues, but does good with emotion, gestures, etc.. The Out-of-distribution generalization is crazy! There are some other systems out there that do it also, but this is good. #ai #deepfake #technology #tech #realtime #beta #microsoft https://lnkd.in/dc2sD2cr
VASA-1 - Microsoft Research
https://www.microsoft.com/en-us/research
-
Wow! Microsoft Research developed VASA-1, an AI technology that can transform a single photo and audio clip into ultra-realistic talking videos in real time! What's groundbreaking is the perfect lip-sync and dynamic expressions that it produces. This is definitely a game-changer. #AI #Innovation #MicrosoftResearch #TechTrends
VASA-1 - Microsoft Research
https://www.microsoft.com/en-us/research
-
"TL;DR: single portrait photo + speech audio = hyper-realistic talking face video with precise lip-audio sync, lifelike facial behaviour, and naturalistic head movements, generated in real time." This is from Microsoft Research. When it goes mainstream, and given a few months more development, you won't be able to believe what you see in video be it online, or on TV. Have a look at this Microsoft link, scroll down and be amazed. #ai #generativeai
VASA-1 - Microsoft Research
https://www.microsoft.com/en-us/research
