Research map / Image generation / Image restoration
FlareReal: A Real-Captured Paired Dataset for Nighttime Lens Flare Removal
Dafeng Zhang, Shizhuo Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Introduces FlareReal, a real-captured paired dataset for nighttime lens flare removal, addressing the limits of synthetic compositing and rendering pipelines.
Problem
Supervised nighttime lens flare removal lacks real flare-corrupted and flare-free image pairs with accurate alignment; existing benchmarks rely on synthetic compositing or rendering that cannot reproduce coupled real-camera effects.
Approach
Capture real paired nighttime images with and without lens flare, aligned accurately, so that the coupled effects of lens contamination, internal reflection, sensor saturation, and automatic exposure are present in the data rather than simulated.
Closest papers by meaning
- Flare7K++: Mixing Synthetic and Real Datasets for Nighttime Flare Removal and Beyond (TPAMI 2024 · similarity 0.95)
- Disentangle Nighttime Lens Flares: Self-supervised Generation-based Lens Flare Removal (AAAI 2025 · similarity 0.93)
- Flare7K: A Phenomenological Nighttime Flare Removal Dataset (NeurIPS 2022 · similarity 0.93)
- FlareX: A Physics-Informed Dataset for Lens Flare Removal via 2D Synthesis and 3D Rendering (NeurIPS 2025 · similarity 0.92)
- Language-Guided One-Step Diffusion Model for Nighttime Flare Removal (CVPR 2026 · similarity 0.90)
- FLARE: Feed-forward Geometry, Appearance and Camera Estimation from Uncalibrated Sparse Views (CVPR 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
