Video-focused generative artificial intelligence startup Haiper received an investment of $13.8 million

Video-focused generative artificial intelligence initiative Haiper, Octopus Ventures received a seed investment of $13.8 million under his leadership. Founded in 2021, Haiper’s management team includes Dr. one of DeepMind’s former researchers. Yishu Miao and Dr. Featuring Ziyu Wang.

The artificial intelligence model that Haiper works on focuses on turning users’ textual commands into a short animated video. Users can also upload images along with written commands to have the model create a creative animation.

We made a way to transform our sketches into video!

The full tutorial on how we did it can be found on our discord:https://t.co/wqcCcN2H8F pic.twitter.com/3hrq5oZWFc

— Haiper AI (@HaiperGenAI) February 23, 2024

It is possible to use Haiper’s application in the beta test version. You can access the model via the company’s website. The site also offers different tools to produce or edit videos. These tools include text-to-video and text-supported image-to-video models.

In fact, so far, besides OpenAI’s Sora model, Meta, Google ve Runway We found that the models produced by have exhibited similar functions. According to the Haiper team, video-focused models are on the verge of a turning point. Artificial intelligence models that focus on video need to increase the quality and accuracy of the videos they create. This requires models to understand the physics context. That means the models in question must understand the interactions between light, motion, texture, and objects. Thus, details such as the way raindrops bounce when hitting different surfaces or the changes in the appearance of objects and people as the wind blows can be depicted more accurately.

Dr. Yishu Miao states that Haiper’s ultimate goal is to develop an artificial general intelligence (AGI) with perceptual capabilities. According to Miao, the company’s visual foundation model will create a leap forward in artificial intelligence’s ability to deeply understand the physics of the world and imitate the essence of reality in the videos it produces.