3rd test with youtube link <iframe width="560" height="315" src="https://lnkd.in/egPzJHGx" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" allowfullscreen></iframe>
Larry Bark’s Post
More Relevant Posts
-
<iframe width="560" height="315" src="https://lnkd.in/esSAjTge" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>
To view or add a comment, sign in
-
<iframe width="560" height="315" src="https://lnkd.in/efqhd9Q6" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" allowfullscreen></iframe>
To view or add a comment, sign in
-
Try the latest video generation model from Google DM😲
Taking a quick dip with our video generation model Veo. 🫧🌊🦭
A quick dip with our generative model Veo
To view or add a comment, sign in
-
<iframe width="560" height="315" src="https://lnkd.in/gK_3TWhX" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" allowfullscreen></iframe>
To view or add a comment, sign in
-
<iframe width="560" height="315" src="https://lnkd.in/dyUtPNHS" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>
To view or add a comment, sign in
-
<iframe width="560" height="315" src="https://lnkd.in/gwkdnTbU" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>
To view or add a comment, sign in
-
<iframe width="560" height="315" src="https://lnkd.in/g96_X-Vc" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>
To view or add a comment, sign in
-
Water wins. As generative video captures "more realism" it is approaching a model of the dynamics of the world, but learning from a perceptual model. "Naive physics" is a term from psychology to describe how we understand the world around us, not from a theory or adult understanding, but from our intuitions and learned experience. We know things fall, or shatter, or crumble, because we know how stones, and wine glasses, and paper behave. Video models are capturing the "naive physics" of visualization but lacking the actual physics. How far will it go? For example, this year hair visualizations are getting a lot better across all models in the field. Generative video models are particularly good at creating water. It makes it acceptable that objects "float" and drift a little more than they should. At some point, our own interval understanding of turbulence, and reflection, "maxes out" and it just looks good and real enough. Unless you stop and look at the details, and then you can see how it's "wrong" or inconsistent with actual physics. But it captured enough of the physical phenomena, that we know what's going on, and looks good. One area of active challenge is the impact of "contact" and "friction" in the visual world. Interestingly, many of these "hard to visualize things" are not crucial for narrative fiction. We can go a long way with what we already have: good models of environments, camera motion, the human body, and expressive talking heads (without showing the feet touching the ground believably, without showing two person interactions) I believe the process of "breaking" these new models will open even more avenues for creative expression.
Taking a quick dip with our video generation model Veo. 🫧🌊🦭
A quick dip with our generative model Veo
To view or add a comment, sign in
-
Wow.... Congratulations Deepmind 👏👏. Almost real life... Unbelievable how fast AI is evolving.... It is now extremely crucial for almost anyone to learn and adapt to AI ASAP. If you don't probably you will end up like a dinosaur. The big question is how many can scale up to an world of AI. If just a few get AI skill set and it becomes a winner take it all scenario ( highly possible..) it will cause social unrest most certainly. Very interesting times ahead. The extreme indebtedness every where makes things even more risky. #artificialintelligence #ai #productivity #macroeconomics #career #future
Taking a quick dip with our video generation model Veo. 🫧🌊🦭
A quick dip with our generative model Veo
To view or add a comment, sign in
-
<iframe width="560" height="315" src="https://lnkd.in/emNxsRMi" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>
To view or add a comment, sign in