META'S AI can turn any text into a video.


Meta announced Make-A-Video, a tool that generates short video clips from text descriptions—an unsettling, albeit inevitable, next step for the world of AI image generation. Watch on YouTube

Generative AI research is pushing creative expression forward by giving people tools to quickly and easily create new content. With just a few words or lines of text, Make-A-Video can bring imagination to life and create one-of-a-kind videos full of vivid colors, characters, and landscapes. The system can also create videos from images or take existing videos and create new ones that are similar.

It's much harder to generate video than photos because beyond correctly generating each pixel, the system also has to predict how they'll change over time. Research Paper

Make-A-Video solves this by adding a layer of unsupervised learning that enables the system to understand motion in the physical world and apply it to traditional text-to-image generation.

The example videos on the Make-A-Video site show videos of “a dog wearing a Superhero outfit with red cape flying through the sky” and “a teddy bear painting a portrait.” The videos are clearly AI-generated, with a blurry, painterly quality native to AI-generated images. Yet, they nonetheless show the fast-moving progress of AI art systems, which only a few years ago were the stuff of memes and science fiction.

Meta seems to be aware of the dangers behind AI art-generating systems, and claims it is “openly sharing this generative AI research and results with the community for their feedback, and will continue to use their responsible AI framework to refine and evolve their approach to this emerging technology.”

But according to the Make-A-Video research paper, the image models were trained using a subset of the LAION dataset, which is known for scraping unfiltered web data that produces biased results.

Within this dataset were images of ISIS executions, nonconsensual nudes, and photoshopped nudes of celebrities. Meta seems to address this issue by parsing down the original data set of over 5.8 billion images down to 2.3 billion, with the paper’s authors claiming, “We filter out sample pairs with NSFW images, toxic words in the text, images with a watermark probability larger than 0.5.

Meanwhile, AI ethics researchers have pushed back against the use of these large language models, warning that their sheer size creates fundamental problems of harmful bias that can not be easily solved. Even Facebook’s own researchers have admitted that their language models have a “high propensity” for producing racist and harmful results.

Comments

Post a Comment

Popular posts from this blog

Scientists Create a Wearable Sensor for Plant Leaves to Detect Water Loss Early

Image
Plants can't speak up when they are thirsty. And visual signs, such as shriveling or browning leaves, don't start until it's too late. Metal electrodes have previously been used to monitor thirsty plants, but it's difficult attaching these devices to hairy leaves, which reduces their accuracy. Now, researchers reporting in ACS Applied Materials & Interfaces have created a wearable sensor for leaves that stays put, and makes it easy to remotely manage plants' water stress. Wearable devices for people are so sophisticated that some smart watches monitor the electrical activity of the wearer's heart, using electrodes that sit against the skin, and send the data to websites, which allow physicians to monitor and assess their patients from a distance. Similarly, plant-wearable devices could help farmers and gardeners remotely monitor plant health, including leaf water content – a key marker of metabolism and drought stress. Renato Lima and colleagues wanted to id
Read more

Google DeepMind is Using AI to Teach Robots to Play Soccer

Image
DeepMind, scientists have successfully trained low-cost humanoid robots to play 1v1 soccer. Using advanced techniques like deep reinforcement learning, the team was able to teach these robots agile skills and tactics, bringing us one step closer to having robots perform complex tasks in the real world.                                                                     Watch on YouTube The training process involved two stages, where robots first learned independent skills like getting up from the ground and scoring goals. In the second stage, these skills were combined into a single soccer-playing agent, which improved its tactics through self-play against previous versions of itself. This progress could lead to the development of advanced humanoid robots capable of performing complex tasks across various industries, such as assisting in disaster relief efforts or navigating intricate environments for search and rescue missions. Additionally, the study explores the potential of multi-a
Read more

Disney introduces a charming, roller-skating mechanical rabbit to help guests "emotionally connect"

Image
  Disney debuted a new robot prototype at SXSW Austin, Texas.  Watch on YouTube Designed to look like the bunny rabbit character Judy Hopps from the movie ‘Zootopia,’ the robot was unveiled by climbing out of a box, rollerblading across the stage and doing somersaults. She’s the size of a little girl, but has big droopy ears like a rabbit, and is wearing roller skates.  On shaky legs, she manages to stand up and regain her balance. With her first moves, it’s obvious she’s already formed an emotional connection with the audience. And with this, the little bunny robot does a front flip and bumps her head before recovering and finding her footing again. She’s then picked up and propped onto her creator’s shoulders like a child.
Read more