I agree. I also wonder how it will deal with movements like sitting down, lying down, running – I imagine it only works well with static scenery.
@fofrai
-
Neo Learning Kung Fu with LoRA Fine-tuning Technique
By
–
I bet when Neo learns Kung Fu he’s using a lora.
-
SVD Impresses with Accurate 3D Image Generation
By
–
Holidays are coming.
— fofr (@fofrAI) 1 décembre 2023
So impressed how accurately SVD has figured out the 3D in this image. The trees, the background buildings, the windows in the gingerbread truck. pic.twitter.com/hRjCE59MaTHolidays are coming. So impressed how accurately SVD has figured out the 3D in this image. The trees, the background buildings, the windows in the gingerbread truck.
-
Diffusers v0.24 Release: IP Adapters, Kandinsky 3.0, Video Diffusion
By
–
Oooh, diffusers v0.24 has a lot of goodies! – IP Adapters
– Kandinsky 3.0
– Stable Diffusion
– SDXL Turbo -
InterpAny-Clearer: Open Source AI Interpretability Project
By
–
Project page is here: https://
zzh-tech.github.io/InterpAny-Clea
rer/
… Github: https://
github.com/zzh-tech/Inter
pAny-Clearer
… -
InterpAny-Clearer Open-Source Video Interpolator Launched on Replicate
By
–
InterpAny-Clearer is a really nice new open-source video interpolator by Zhihang Zhong et al.
— fofr (@fofrAI) 29 novembre 2023
It's now up on Replicate, courtesy of @lucataco93, perfect for interpolating SVD.https://t.co/aEcVnQMHDO
Here's an example:https://t.co/hDAmOPINxA pic.twitter.com/suS7hxiAKSInterpAny-Clearer is a really nice new open-source video interpolator by Zhihang Zhong et al. It's now up on Replicate, courtesy of @lucataco93, perfect for interpolating SVD. https://
replicate.com/lucataco/inter
pany-clearer
… Here's an example: https://
replicate.com/p/nlmwvtlbhbtg
j3xx5ae5633c3m
… -
SDXL Turbo image quality issues with multiple steps
By
–
Has anyone else had issues with SDXL Turbo giving bad images with steps > 1? Maybe it's the scheduler I'm using?
-
Fine-tuning and derivatives of core AI models
By
–
What the situation with derivatives of core models, like fine tunes for example?
-
GPU Memory Limits for Large Model Training on A100
By
–
No I’ve gone higher, 2k would OOM error on an 80gb A100