Tech giant Meta is reportedly working on an Artificial Intelligence (AI) model known as ImageBind, that will imitate human perception in so many areas.
According to reports, ImageBind is part of Meta’s initiative to create multimodal systems that can learn from different types of data. Engineers at Meta while creating the AI tool adopted a multi-sensory framework that involves audio, text, video, images, etc.
The AI model not only understands an element but is capable of linking it with other features. For example, it will be able to determine temperature, sound, and shape, among others.
Speaking on the creation of the AI model, Meta said,
“In typical AI systems, there is a specific embedding that is vectors of numbers that can represent data and their relationships in machine learning for each respective modality. ImageBind shows that it is possible to create a joint embedding space across multiple modalities without needing to train on data with every different combination of modalities.
“This is important because it is not feasible for researchers to create datasets with samples that contain, for example, audio data and thermal data from a busy City Street, or depth data and a text description of a seaside cliff. While we explored six modalities in our current research, we believe that introducing new modalities that link as many senses as possible, like touch, smell, speech, and brain fMRI signals, will enable richer human-centric AI models”.
Investors King understands that in 2016, the CEO of Facebook Mark Zuckerberg, had stated that AI advancements will give machines sensory capabilities that are superior to humans within five to 10 years.
According to Zuckerberg, he predicts that there would be Artificial Intelligence machines with human-level perception within the 2021 to 2026 timeframe. He believes that when this happens, humans will benefit from a lot better experiences, as he puts it, such as a better news feed on the Facebook app.