Hey, Nadim here, back with another AI video! This time, I will be exploring a new AI video generator called "Vidu" that claims to be on par with OpenAI's Sora.
Vidu AI Website - https://www.shengshu-ai.com/home
Veo: New AI Video Model from Google - • New AI Video Tool from Google - Veo! SORA ...
#aivideogenerator #sora #texttovideoai
Vidu is created by a Chinese startup named “Shengshu Technology” in collaboration with Tsinghua University. It's capable of generating text-to-video up to 16 seconds long at Full HD resolution.
What's really exciting about the videos generated with “Vidu” is that they show some of the best demonstrations of coherence, consistency, and understanding of the real world that we've seen. So, we're getting a lot less strange limbs morphing or people not actually interacting in reality in a sensible way.
Now, how does Vidu match up to Sora? That's the real question! Vidu has showcased some videos that are a direct comparison to the videos generated with OpenAI’s Sora.
And looking at these videos side by side, I would say that Vidu is doing really well. It definitely outcompetes all the video generators out there that we can use today, including Runway and Pika.
However, I am not sure if videos generated with Vidu are on par with Sora. Sora's result does seem to be of higher quality.