Reproduce “UBLLIE: Unified Backlight and Low-Light Image Enhancement” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yasmin Yasin, Muhammad Usman, Ibrahim Radwan, Saeed Anwar
UBLLIE is an unsupervised image enhancement framework developed by researchers from King Fahd University of Petroleum and Minerals, Ontario Tech University, University of Western Australia, and University of Canberra. It addresses both backlit and low-light conditions by integrating CLIP-guided semantic supervision with a U-Net and Atrous Spatial Pyramid Pooling (ASPP) architecture, leading to improved visual quality and detail recovery without requiring paired ground-truth data.
Key insights
- Leveraging high-level semantic understanding from pre-trained vision-language models like CLIP can provide effective unsupervised guidance for image enhancement, reducing the dependency on explicit pixel-level ground truth.
- A carefully designed network architecture, combining residual U-Net with ASPP, can effectively capture multi-scale contextual information to handle spatially heterogeneous illumination degradations, leading to nuanced and artifact-free corrections.
- Unsupervised methods can achieve competitive or superior quantitative and qualitative results compared to supervised baselines in both backlit and low-light conditions, offering a more practical and generalizable solution for real-world scenarios.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.