Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priors · HackerPapers discussions

Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priors — mentioned by deviparikh

Thu, 14 Jul 2022 15:21:21 GMT

Blog post: https://ai.facebook.com/blog/greater-creative-control-for-ai... Paper: https://arxiv.org/abs/2203.13131 Meta AI's tweet: https://twitter.com/MetaAI/status/1547598454922153985?s=20&t...

Paper and discussion history. A mention is not an endorsement.

Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priors — mentioned by ht96

Thu, 14 Jul 2022 15:17:32 GMT

More examples here https://www.instagram.com/p/Cf_tjL_rgEe/?hl=en I understand this is probably just a frontend to this paper from March https://arxiv.org/pdf/2203.13131.pdf Given the spotlight DALL-E has received makes sense to capitalise

Paper and discussion history. A mention is not an endorsement.

Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priors — mentioned by gwern

Thu, 19 May 2022 01:26:19 GMT

Facebook's Make-A-Scene, which looks almost as good as DALL-E 2, does something very much like that, using face-specific targeting to boost the quality of that part of images: https://arxiv.org/abs/2203.13131#facebook

Paper and discussion history. A mention is not an endorsement.

Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priors — mentioned by gwern

Wed, 13 Apr 2022 16:43:08 GMT

Also Make-A-Scene, which in some ways is noticeably better than DALL-E 2 (faces, editing & control of layout through semantic segmentation conditioning): https://arxiv.org/abs/2203.13131#facebook

Paper and discussion history. A mention is not an endorsement.

Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priors — mentioned by astrange

Wed, 13 Apr 2022 02:35:49 GMT

> What is the content of the training set like, that DALL·E 2 has all this info? Web crawling. (https://arxiv.org/pdf/2103.00020.pdf 2.2) > Are there people out there spending thousands of hours just tagging or writing descriptions of training images? "Yes", Google Image Search, Wikipedia captions, Danbooru are all that in some sense. The biases should be obvious if you think about it for a minute and CLIP has even more biases since they removed all kinds of NSFW content, etc. > And does DALL·E just receive the image as-is with its description, or does it get more info (e.g. "this part of the…

Paper and discussion history. A mention is not an endorsement.

Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priors — mentioned by gwern

Mon, 11 Apr 2022 17:48:56 GMT

It's the PR risk. You can easily pass the 'look good to an uniformed observer' test with human faces. Remember, faces were something GANs were doing near-flawlessly back as far as ProGAN all the way back in the dark ages of late 2017. (Then StyleGAN did non-photograph faces, like anime - see my ThisWaifuDoesNotExist for a demo of that.) Doing them as part of a larger composition, where the face is a small part of the images and may vary much more than in closeup centered portraits, is harder, but take a look at how Facebook's DALL-E rival Make-A-Scene does it…

Paper and discussion history. A mention is not an endorsement.

Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priors — mentioned by gwern

Thu, 07 Apr 2022 03:09:20 GMT

As I keep telling people: "hands are hard". This is why I went so far as to make a hand-specific dataset ("PALM" https://www.gwern.net/Crops#palm which of course now everyone is going to confuse with 'PaLM'...). Hands are just way too variable to learn easily. My dataset is a start, but it may benefit from focused training, the way Facebook's new Make-A-Scene https://arxiv.org/abs/2203.13131#facebook (not DALL-E 2 quality but not far from it) has focused losses on faces.

Paper and discussion history. A mention is not an endorsement.

Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priors — mentioned by smusamashah

Fri, 25 Mar 2022 08:11:50 GMT

As HTML https://www.arxiv-vanity.com/papers/2203.13131/

Paper and discussion history. A mention is not an endorsement.

Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priors — HN submission

Fri, 25 Mar 2022 08:11:16 GMT

Make-a-Scene: Scene-Based Text-to-Image Generation with Human Priors

2 points; 1 discussion reply on HN. Paper and source history. A submission is not an endorsement.