ICCP 2025 Paper: Unified Model for Joint Demosaicing on New Smartphone Sensors
CSProfKGD · x · 2026-08-12
A new paper at ICCP 2025 discusses strategies for joint demosaicing and denoising for new smartphone sensors (Single-, Quad-, and Nona-Bayer patterns).
Key Findings & Methods:
- Existing deep network-based models typically train a separate network per pattern, requiring extra memory overhead and making camera switching difficult.
- The study found that concatenating a three-channel mosaic embedding to the input image and training a unified demosaicing architecture outperforms existing Quad- and Nona-Bayer models and is comparable to Single-Bayer models.
- The authors describe a maskout strategy that enhances model performance and facilitates dead pixel correction, a step often overlooked by existing AI-based models.
- A new demosaicing dataset was captured and released as part of the effort.
More from Research
- Codex Falsely Reports Success 4.1% of the Time, Open-Source Tool Reveals — Due_Emu_8229 · 2026-08-12
- Kimi K3 Distillation Controversy: Authors Admit No Proof, Likely Data Contamination — bookwormengr · 2026-08-12
- RefineAny3D: Using fine-tuned VLMs to optimize monocular 3D detection — kwangmoo_yi · 2026-08-12
- Tabular ML Question: When to Model Categorical Variables Separately? — mariofilhoml · 2026-08-12
- Frontier LLMs Lack Theory of Mind, Developer Calls for Targeted RL Training — lateinteraction · 2026-08-12
- MLS-Bench: AI Agents Can Optimize ML Experiments But Fail to Discover New Methods — 机器之心 · 2026-08-12