Facebook Introduces “Anticipative Video Transformer”

Here is a new offering from Facebook’s machine learning. Facebook introduces “Anticipative Video Transformer (AVT).” It predicts future actions based on visual interpretations. Facebook uses machine learning to personalized ads. The platform has automated campaign ideas generation and consumer targeting based on their shopping habits. All of these aimed to enhance eCommerce solutions.

Overview of Facebook's visual anticipation process

Facebook’s Anticipative Video Transformer (AVT) is an end-to-end model of anticipating the next actions on videos. They can be especially useful for artificial intelligence (AI) assistance. Based on previous interactions, AVT can react ahead of time or warn someone of a possible mistake. The feature is not something that the platform will roll out soon. But they are looking forward to incorporating the function on their AR/VR devices shortly.

Facebook introduces “Anticipative Video Transformer” on 13 October 2021.

Implications for Marketers: 

Facebook’s “Anticipative Video Transformer” is a new way to anticipate actions and prevent possible mistakes. For marketers, they can be a key tool in offering consumer-facing products to guide people in a variety of tasks.

Reference: https://ai.facebook.com/blog/anticipative-video-transformer-improving-ais-ability-to-predict-whats-next-in-a-video

If you find this post useful, please share to your friends.

Join Our Newsletter

Join our newsletter to receive the latest updates and market tips.