Ismael's picture

Ismael

IsmaelMousa

AI & ML interests

NLP

Recent Activity

liked a model 25 days ago
Qwen/Qwen2.5-32B
liked a model about 1 month ago
FacebookAI/roberta-base
View all activity

Organizations

Open-Source AI Meetup's profile picture OpenOrca's profile picture MLX Community's profile picture INNOVA AI's profile picture Narra's profile picture llmc's profile picture

IsmaelMousa's activity

New activity in IsmaelMousa/movies 4 months ago
New activity in IsmaelMousa/books 4 months ago
updated a Space 4 months ago
reacted to DmitryRyumin's post with šŸ”„ 6 months ago
view post
Post
3642
šŸš€šŸŽ­šŸŒŸ New Research Alert - Portrait4D-v2 (Avatars Collection)! šŸŒŸšŸŽ­šŸš€
šŸ“„ Title: Portrait4D-v2: Pseudo Multi-View Data Creates Better 4D Head Synthesizer šŸ”

šŸ“ Description: Portrait4D-v2 is a novel method for one-shot 4D head avatar synthesis using pseudo multi-view videos and a vision transformer backbone, achieving superior performance without relying on 3DMM reconstruction.

šŸ‘„ Authors: Yu Deng, Duomin Wang, and Baoyuan Wang

šŸ“„ Paper: Portrait4D-v2: Pseudo Multi-View Data Creates Better 4D Head Synthesizer (2403.13570)

šŸŒ GitHub Page: https://yudeng.github.io/Portrait4D-v2/
šŸ“ Repository: https://github.com/YuDeng/Portrait-4D

šŸ“ŗ Video: https://www.youtube.com/watch?v=5YJY6-wcOJo

šŸš€ CVPR-2023-24-Papers: https://github.com/DmitryRyumin/CVPR-2023-24-Papers

šŸ“š More Papers: more cutting-edge research presented at other conferences in the DmitryRyumin/NewEraAI-Papers curated by @DmitryRyumin

šŸš€ Added to the Avatars Collection: DmitryRyumin/avatars-65df37cdf81fec13d4dbac36

šŸ” Keywords: Portrait4D #4DAvatar #HeadSynthesis #3DModeling #TechInnovation #DeepLearning #ComputerGraphics #ComputerVision #Innovation
  • 1 reply
Ā·
reacted to merve's post with šŸ¤— 6 months ago
view post
Post
6044
Fine-tune Florence-2 on any task šŸ”„

Today we release a notebook and a walkthrough blog on fine-tuning Florence-2 on DocVQA dataset @andito @SkalskiP

Blog: https://huggingface.co/blog šŸ“•
Notebook: https://colab.research.google.com/drive/1hKDrJ5AH_o7I95PtZ9__VlCTNAo1Gjpf?usp=sharing šŸ“–
Florence-2 is a great vision-language model thanks to it's massive dataset and small size!

This model requires conditioning through task prefixes and it's not as generalist, requiring fine-tuning on a new task, such as DocVQA šŸ“

We have fine-tuned the model on A100 (and one can also use a smaller GPU with smaller batch size) and saw that model picks up new tasks šŸ„¹

See below how it looks like before and after FT šŸ¤©
Play with the demo here andito/Florence-2-DocVQA šŸ„ā€ā™€ļø
upvoted 2 articles 7 months ago
view article
Article

Falcon 2: An 11B parameter pretrained language model and VLM, trained on over 5000B tokens tokens and 11 languages

ā€¢ 25
view article
Article

Introducing the Open Arabic LLM Leaderboard

ā€¢ 76
upvoted an article 8 months ago
view article
Article

Hugging Face x LangChain : A new partner package in LangChain

ā€¢ 114