tg-me.com/DeepLearning_ai/1151
Last Update:
ππ Magic-Me: Identity-Specific Video ππ
πhashtag#ByteDance (+UC Berkeley) unveils VCD for video-gen: with just a few images of a specific identity it can generate temporal consistent videos aligned with the given prompt. Impressive results, source code under Apache 2.0 π
ππ’π π‘π₯π’π π‘ππ¬:
β
Novel Video Custom Diffusion (VCD) framework
β
High-Quality ID-specific videos generation
β
Improvement in aligning IDs-images and text
β
Robust 3D Gaussian Noise Prior for denoising
β
Better Inter-frame correlation / video consistency
β
New modules F-VCD/T-VCD for videos upscale
β
New train with masked loss by prompt-to-segmentation
hashtag#artificialintelligence hashtag#machinelearning hashtag#ml hashtag#AI hashtag#deeplearning hashtag#computervision hashtag#AIwithPapers hashtag#metaverse
πChannel: @deeplearning_ai
πPaper https://arxiv.org/pdf/2402.09368.pdf
πProject https://magic-me-webpage.github.io/
πCode https://github.com/Zhen-Dong/Magic-Me
BY Artificial Intelligence && Deep Learning
Warning: Undefined variable $i in /var/www/tg-me/post.php on line 280
Share with your friend now:
tg-me.com/DeepLearning_ai/1151