All-Optical Image Identification with Programmable Matrix Transformation

Baohua Ni; Fang Liu; Kaiyu Cui; Shikang Li; Wei Zhang; Xue Feng; Yidong Huang

All-Optical Image Identification with Programmable Matrix Transformation

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2104.02474 v1 pith:USOPY4UB submitted 2021-04-01 physics.optics eess.IV

Shikang Li , Baohua Ni , Xue Feng , Kaiyu Cui , Fang Liu , Wei Zhang , Yidong Huang This is my paper

classification physics.opticseess.IV

keywords matrixprogrammableall-opticalarchitecturehighopticalproposedspatial

verification ladder T0 review T1 audit T2 compute T3 formal

0 comments

read the original abstract

An optical neural network is proposed and demonstrated with programmable matrix transformation and nonlinear activation function of photodetection (square-law detection). Based on discrete phase-coherent spatial modes, the dimensionality of programmable optical matrix operations is 30~37, which is implemented by spatial light modulators. With this architecture, all-optical classification tasks of handwritten digits, objects and depth images are performed on the same platform with high accuracy. Due to the parallel nature of matrix multiplication, the processing speed of our proposed architecture is potentially as high as7.4T~74T FLOPs per second (with 10~100GHz detector)

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

Experimental investigation of single qubit quantum classifier with small number of samples
quant-ph 2025-07 conditional novelty 4.0 of 10

A silicon photonic single-qubit classifier achieves about 86 percent accuracy when trained with an average of roughly two photons per sample, matching simulation.

Pith tools