#437 · Primary category： Computer Vision
FBCNN
Official Code for ICCV 2021 paper "Towards Flexible Blind JPEG Artifacts Removal (FBCNN)"
Project last updated：04/19/24
GitHub Stars
529
Forks
51
Contributors
2
License
Apache-2.0
Why we included this project
JPEG artifacts are a constant nuisance in image processing, and they get harder to remove when you do not know what quality setting produced them. FBCNN is an official ICCV 2021 implementation that removes blockiness and ringing from JPEG images without needing the quality factor up front, so it copes with re-compressed files where that information is missing. It also handles double JPEG compression, aligned or not, a case that breaks simpler deblocking methods. The repo ships pretrained models and a Gradio demo, so you can try your own photos before wiring it into a pipeline. For teams cleaning images before deeper analysis or restoration, that blind design is the real payoff: nothing to tune per source.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
opencv
Open Source Computer Vision Library
RuView
π RuView turns commodity WiFi signals into real-time spatial intelligence, vital sign monitoring, and presence detection — all without a single pixel of video.
PaddleOCR
Turn any PDF or image document into structured data for your AI. A powerful, lightweight OCR toolkit that bridges the gap between images/PDFs and LLMs. Supports 100+ languages.
MinerU
Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your Agentic workflows.
tesseract
Tesseract Open Source OCR Engine (main repository)