A 128×96 Multimodal Flash LiDAR SPAD Imager with Object Segmentation Latency of 18μs Based on Compute-Near-Sensor Ising Annealing Machine
Abstract
This paper presents a$128 \times 96$multimodal flash LiDAR SPAD imager integrated with a compute-near-sensor Ising-model annealing processor for dynamic object segmentation, framed as a combinatorial optimization problem (COP). The system enables real-time updates of Hamiltonian coefficients using a high-bandwidth output-while-write scheme, and accelerates Ising spin iterations by exploiting multimodal SPAD information, achieving a segmentation latency of$18 \mu \mathrm{s}$.
Fields of Study & Subjects
Advanced Optical Sensing Technologies
Advanced Neural Network Applications
Advanced Memory and Neural Computing
Publication details
- Journal / source
- Digest of technical papers/Digest of technical papers - IEEE International Solid-State Circuits Conference
- Pages
- 122–124
- Publication date
- 15 February 2026
- Language
- EN