Thank you for the explanation. It does seem like you mixed up the diffusion processes though. At 2:50 you refer to the process of adding noise as the backwards process, and removing noise as the forward process. In section 2 of the DDPM paper, they define the reverse process as starting with noise, and the forward process as the one where Gaussian noise is added.
Thanks! I originally thought that just a historical retelling might be a bit dry, but in this case the new ideas clearly developed as the topic ping-ponged between several research labs. So the explanation, coupled with the 'race', seemed like a good fit :-)
@@MartinAndrews-mdda Do you plan on making a follow-up on this? With ControlNet, LoRA, etc. Also, is there a place where I can watch your other MLSG presentations?
Great video, thanks for sharing this talk. Being able to tile images using stable diffusion is not a niche function, it's incredibly useful for artists generating textures for 3d models. This will be one of the biggest use cases for this tech in it's current state. Also that's probably not a pixie... that's Link from the Legend of Zelda, you're obviously not a gamer lol.
Only gaming I do is Minecraft :-) So perhaps I shouldn't have made such a big deal of how 'niche' the tileable thing was... Good catch identifying the pixie!
Very clear explanation, I recommended your channels to friends who want to catch up with recent developments in AI.
Very interesting and well explained, thanks a lot for your efforts
Thanks a lot, the talk covered all the topics i wanted to know.
Excellent coverage of a large topic. Cheers
Thank you for the explanation. It does seem like you mixed up the diffusion processes though. At 2:50 you refer to the process of adding noise as the backwards process, and removing noise as the forward process. In section 2 of the DDPM paper, they define the reverse process as starting with noise, and the forward process as the one where Gaussian noise is added.
Martin. Great job organizing this fast changing area!
Thanks! I originally thought that just a historical retelling might be a bit dry, but in this case the new ideas clearly developed as the topic ping-ponged between several research labs. So the explanation, coupled with the 'race', seemed like a good fit :-)
@@MartinAndrews-mdda you did so well sir. It's not easy to be so clear and concise when explaining such a complex space. You are very talented
Amazing video!
Thank you so much for taking your time to do it!
Great comment! Thank you so much for taking your time to make it! Genuinely appreciated :-)
@@MartinAndrews-mdda Do you plan on making a follow-up on this? With ControlNet, LoRA, etc.
Also, is there a place where I can watch your other MLSG presentations?
Thank you so much! I am getting into stable diffusion, I learned so much about it.
Very nice informative vid, and for once a teacher who speaks clearly and quietly 👍
Great video, thanks for sharing this talk. Being able to tile images using stable diffusion is not a niche function, it's incredibly useful for artists generating textures for 3d models. This will be one of the biggest use cases for this tech in it's current state. Also that's probably not a pixie... that's Link from the Legend of Zelda, you're obviously not a gamer lol.
Only gaming I do is Minecraft :-) So perhaps I shouldn't have made such a big deal of how 'niche' the tileable thing was... Good catch identifying the pixie!