Introducing Make-A-Video3D! Generating 3D dynamic (mini) scenes from input text.
That is, text --> 4D!
Needs no 4D data (i.e., no dynamic 3D data), no static 3D data, no paired text-video data.
Paper: arxiv.org/abs/2301.11280
Website: make-a-video3d.github.io/
I spent a couple months on and off working in Tokyo for Mitsubishi and I find myself daydreaming of it often.
@liamwongβs photography is the only physical thing Iβve found that matches those faded memories.
His new book is a buy π
Liam Wong: After Dark a.co/d/74d8yVm
This is the only real way to respond to AI - experiment, discover, and adapt. Pretending that new technology can be made to disappear isnβt realistic. If you donβt, someone else will.
If there is one clear lesson about new general purpose technologies, it is that they will be used in unexpected ways.
Within just a couple weeks, AI was already being used by my students to solve all sorts of novel problems. I wrote about what that means. oneusefulthing.substack.com/β¦
Fresh mint on @TeiaArt πΈοΈ
~no one is there
"He is calling and throwing his arms up in the air, and no one is there."
Camera Feedback Panasonic WJ-MX30 Edirolv-4.
Edition of 20 available for 3 tez π«
If you read another doom & gloom piece on AI art, please make sure to watch this, too, because it's an example of sober, well-researched reporting (by #776Titan@cleoabram)
youtube.com/watch?v=NiJeB2NJβ¦
This is an interesting perspective as to why what works on the page doesnβt always work on screen. Also appreciate the Prince of Tides shoutout: at.tumblr.com/filmcourage/whβ¦