Researchers have developed a unified framework for understanding the fundamental recovery limits of single-photon avalanche diodes (SPADs). The paper derives the likelihood score function for various SPAD operation modes, which forms the basis for both theoretical recovery bounds and practical reconstruction algorithms. This unified approach reveals a significant performance gap between different modes, with timestamped modes degrading only linearly compared to the exponential saturation of binary counts. The work also extends diffusion posterior sampling to timestamped SPAD data, demonstrating that aligning the score function with the operation mode enhances reconstruction fidelity. AI
IMPACT Establishes theoretical limits and practical methods for single-photon sensing, potentially impacting future sensor design and data analysis.
RANK_REASON Academic paper detailing new theoretical framework and algorithms for signal processing. [lever_c_demoted from research: ic=1 ai=0.4]
- arXiv
- Bayesian Cramer-Rao lower bounds
- computer science
- Computer vision and pattern recognition
- Lior Dvir
- Single-Photon Avalanche Diodes for Cancer Diagnosis
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
