Even_Adder

joined 1 year ago
[–] [email protected] 2 points 2 weeks ago

Here's a video explaining how diffusion models work, and this article by Kit Walsh, a senior staff attorney at the EFF.

[–] [email protected] 2 points 2 weeks ago

Your comment made my day. Thanks.

[–] [email protected] 0 points 2 weeks ago (9 children)

Anyone spreading this misinformation and trying gatekeep being an artist after the avant-garde movement doesn't have an ounce of education in art history. Generative art, warts and all, is a vital new form of art that's shaking things up, challenging preconceptions, and getting people angry - just like art should.

[–] [email protected] 5 points 3 weeks ago

Entertainment.

[–] [email protected] 4 points 4 weeks ago

Their policy could never stop anyone in the first place.

[–] [email protected] 6 points 1 month ago

Using copyrighted works without permission isn't illegal and shouldn't be. You should check out this article by Kit Walsh, a senior staff attorney at the EFF, and this open letter by Katherine Klosek, the director of information policy and federal relations at the Association of Research Libraries.

[–] [email protected] 31 points 1 month ago* (last edited 1 month ago) (2 children)

Someone dumb enough could easily flatten someone backing up with that bug.

[–] [email protected] 6 points 2 months ago (1 children)

Or just not show people what you're typing.

[–] [email protected] 7 points 2 months ago (1 children)

I can't tell if this is a joke or not.

[–] [email protected] 4 points 2 months ago

A computer like that is useful outside of work. I'd pay for it out of pocket if I had to.

[–] [email protected] 7 points 2 months ago

The only thing I got from this is that bro loves ads more than anything in the world.

[–] [email protected] 4 points 2 months ago

I accept regulations are real, but not all ways to help people require you dealing with regulations. I'm still waiting on that proof by the way.

23
submitted 1 year ago* (last edited 1 year ago) by [email protected] to c/[email protected]
 

Abstract:

Significant advancements have been achieved in the realm of large-scale pre-trained text-to-video Diffusion Models (VDMs). However, previous methods either rely solely on pixel-based VDMs, which come with high computational costs, or on latent-based VDMs, which often struggle with precise text-video alignment. In this paper, we are the first to propose a hybrid model, dubbed as Show-1, which marries pixel-based and latent-based VDMs for text-to-video generation. Our model first uses pixel-based VDMs to produce a low-resolution video of strong text-video correlation. After that, we propose a novel expert translation method that employs the latent-based VDMs to further upsample the low-resolution video to high resolution. Compared to latent VDMs, Show-1 can produce high-quality videos of precise text-video alignment; Compared to pixel VDMs, Show-1 is much more efficient (GPU memory usage during inference is 15G vs 72G). We also validate our model on standard video generation benchmarks. Our code and model weights are publicly available at https://github.com/showlab/Show-1.

view more: next ›