I'm a fashion and editorial photographer, and a mixed media artist.
This channel is dedicated to generative AI R&D, applied to fashion photography and other creative fields. Tutorials, research, random projects and more.
Long time no see!
I'll be back at the end of the week with a in-depth video about training custom VTON models, why I disappeared for a while, and what I learnt in the meantime.
So if you're looking for a long, boring, and nerdy explanation of virtual try ons, you know where to find me.
See you soon!
Next video is coming up on Tuesday or Wednesday because I think I'm onto something kinda stupid but also kinda fun, but I need more time to test it. Sorry for the delay!
I know most of the ones who reach out to me for support don't read these posts, but I'll write this just in case. Earlier this month I started working as Head of Creative AI, and between that and the other projects / clients I'm overseeing, it leaves me very limited time to support my past YouTube tutorials. It's hard enough to keep my video schedule to one video per week, if I were to actively support all my workflows I wouldn't have the time to work, let alone to live.
I have noticed a recent uptick in requests for one on one support, and I'm sorry, but I simply don't have the time to fix single user issues, or guide you through mostly ENV or OS-dependent issues. Python is very clear in describing the runtime errors it encounters, and the secret to how I build my workflows is... I encounter these errors too! I just google the issues if they're not clear enough and try to find a solution.
For example, there's been a Adjustment Level issue on my first Frequency Separation workflows, that it's due to a node update on the pack I was using. It's a simple fix for me, but it takes time to: - find all the workflows that use that logic - find the correct values - implement changes - deploy changes There are already multiple solutions posted by different users on how to solve that issue, both on the videos and on OpenArt. I will solve that issue myself when I'll have the time to, but please don't expect a timeline.
I strive for keeping my daily life separate from my YouTube life, but please remember: I am a human being first, a YouTube creator second. I try to reply to all the emails I get, but please understand that there's a limit to how much I can get done in a day, and you're most probably not the only one reaching out.
30 mins video today on Flux LoRA training, with extras such as single image training and target blocks training. It's a long one, but I go from the basics to advanced stuff.
I had a flu over the past few days, so no new video on Monday :( I'll try to record one and publish it during the week, but it all depends on how much free time I'll have. Sorry about that :/
Today's video, a Flux review from a professional pov, will be out slightly later than usual - around 3.30pm. It's a bit long and I was very busy with deadlines all week. Sorry about that!
Product Video Relight tomorrow: moving subject + moving light, moving subject + "still" light, still subject + moving light, still subject + "still" light. Without AnimateDiff, with color matching and detail preservation. See y'all at 2.30pm CET!
Andrea Baioni
Long time no see!
I'll be back at the end of the week with a in-depth video about training custom VTON models, why I disappeared for a while, and what I learnt in the meantime.
So if you're looking for a long, boring, and nerdy explanation of virtual try ons, you know where to find me.
See you soon!
6 months ago | [YT] | 53
View 7 replies
Andrea Baioni
I'm back, what did I miss? :P
Sorry for the unforeseen wait, life happened.
1 year ago | [YT] | 16
View 5 replies
Andrea Baioni
Next video is coming up on Tuesday or Wednesday because I think I'm onto something kinda stupid but also kinda fun, but I need more time to test it. Sorry for the delay!
1 year ago | [YT] | 18
View 1 reply
Andrea Baioni
I know most of the ones who reach out to me for support don't read these posts, but I'll write this just in case.
Earlier this month I started working as Head of Creative AI, and between that and the other projects / clients I'm overseeing, it leaves me very limited time to support my past YouTube tutorials. It's hard enough to keep my video schedule to one video per week, if I were to actively support all my workflows I wouldn't have the time to work, let alone to live.
I have noticed a recent uptick in requests for one on one support, and I'm sorry, but I simply don't have the time to fix single user issues, or guide you through mostly ENV or OS-dependent issues.
Python is very clear in describing the runtime errors it encounters, and the secret to how I build my workflows is... I encounter these errors too! I just google the issues if they're not clear enough and try to find a solution.
For example, there's been a Adjustment Level issue on my first Frequency Separation workflows, that it's due to a node update on the pack I was using. It's a simple fix for me, but it takes time to:
- find all the workflows that use that logic
- find the correct values
- implement changes
- deploy changes
There are already multiple solutions posted by different users on how to solve that issue, both on the videos and on OpenArt.
I will solve that issue myself when I'll have the time to, but please don't expect a timeline.
I strive for keeping my daily life separate from my YouTube life, but please remember: I am a human being first, a YouTube creator second. I try to reply to all the emails I get, but please understand that there's a limit to how much I can get done in a day, and you're most probably not the only one reaching out.
1 year ago (edited) | [YT] | 58
View 12 replies
Andrea Baioni
Z-Depth Animations (Houdini-Like) today! I'll upload the video at 3:30pm CET.
See you then!
1 year ago | [YT] | 41
View 0 replies
Andrea Baioni
Sorry, no video this week because I have been out of office without access to recording equipment.
Maybe we'll do a live later this week?
1 year ago | [YT] | 15
View 1 reply
Andrea Baioni
30 mins video today on Flux LoRA training, with extras such as single image training and target blocks training.
It's a long one, but I go from the basics to advanced stuff.
1 year ago | [YT] | 27
View 0 replies
Andrea Baioni
I had a flu over the past few days, so no new video on Monday :(
I'll try to record one and publish it during the week, but it all depends on how much free time I'll have.
Sorry about that :/
1 year ago | [YT] | 26
View 4 replies
Andrea Baioni
Today's video, a Flux review from a professional pov, will be out slightly later than usual - around 3.30pm. It's a bit long and I was very busy with deadlines all week. Sorry about that!
1 year ago | [YT] | 16
View 1 reply
Andrea Baioni
Product Video Relight tomorrow: moving subject + moving light, moving subject + "still" light, still subject + moving light, still subject + "still" light.
Without AnimateDiff, with color matching and detail preservation. See y'all at 2.30pm CET!
1 year ago | [YT] | 36
View 0 replies
Load more