Stars
Using Chord DHT to build the file shared system with Java
The system is a P2P network leveraging a DHT for efficient file sharing. A central server manages a Universal Hashed Peer Table (UHPT) with peers' GUID, IP, port, and Routing Metric, and a Univers…
[ACM MM 2021] Visible Watermark Removal via Self-calibrated Localization and Background Refinement
[ACM MM 2021] Visible Watermark Removal via Self-calibrated Localization and Background Refinement
hch-NLP / REMOVE-WATERMARKS-FROM-IMAGES-WITH-GANS
Forked from Pruthvi-Radadiya/REMOVE-WATERMARKS-FROM-IMAGES-WITH-GANSPix2Pix GAN architecture is implemented to remove watermarks from images without marking the position of watermark.
Pix2Pix GAN architecture is implemented to remove watermarks from images without marking the position of watermark.
Project for detecting and removing visible watermarks for images
Project for detecting and removing visible watermarks for images
Using depth learning to remove watermark from image
Python script to remove watermark from any image with watermark in it.
Python script to remove watermark from any image with watermark in it.
hch-NLP / DDNM
Forked from wyhuai/DDNM[ICLR 2023 Oral] Zero-Shot Image Restoration Using Denoising Diffusion Null-Space Model
[ICLR 2023 Oral] Zero-Shot Image Restoration Using Denoising Diffusion Null-Space Model
Real-ESRGAN aims at developing Practical Algorithms for General Image/Video Restoration.
[ECCV 2024] codes of DiffBIR: Towards Blind Image Restoration with Generative Diffusion Prior
End-to-end model training and deployment reference for handwritten Chinese text recognition, and can also be extended to other languages.
End-to-end model training and deployment reference for handwritten Chinese text recognition, and can also be extended to other languages.
An image captioner with Chinese language
An image captioner with Chinese language
Train a model for Image Caption from ViT and GPT pretrained model
基于ClipCap的看图说话Image Caption模型
多模态中文LLaMA&Alpaca大语言模型（VisualCLA）
hch-NLP / LLaSM
Forked from LinkSoul-AI/LLaSM第一个支持中英文双语语音-文本多模态对话的开源可商用对话模型。便捷的语音输入将大幅改善以文本为输入的大模型的使用体验，同时避免了基于 ASR 解决方案的繁琐流程以及可能引入的错误。
第一个支持中英文双语语音-文本多模态对话的开源可商用对话模型。便捷的语音输入将大幅改善以文本为输入的大模型的使用体验，同时避免了基于 ASR 解决方案的繁琐流程以及可能引入的错误。
多模态中文LLaMA&Alpaca大语言模型（VisualCLA）
Awesome Pretrained Chinese NLP Models，高质量中文预训练模型&大模型&多模态模型&大语言模型集合
hch-NLP / LAVIS
Forked from salesforce/LAVISLAVIS - A One-stop Library for Language-Vision Intelligence