Related Experiment Video
Updated: Jan 11, 2026
11:18
Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks
Published on: March 2, 2015
10.8K
Spiking Neural Networks in Imaging: A Review and Case Study
Michael Voudaskas1,2, Jack Iain MacLean1, Neale A W Dutton2
1Institute for Integrated Micro and Nano Systems, The University of Edinburgh, Edinburgh EH9 3BF, UK.
Sensors (Basel, Switzerland)
|November 13, 2025
Summary
Spiking neural networks (SNNs) show potential for energy-efficient imaging but face challenges with datasets, training, and hardware integration. Future work needs benchmarks and hardware-aware training for broader applications.
Area of Science:
- Artificial Intelligence
- Computer Vision
- Neuroscience
Background:
- Spiking neural networks (SNNs) offer energy-efficient, event-driven computation.
- SNNs are being explored for various imaging applications.
Purpose of the Study:
- To review the current state of SNNs in imaging.
- To identify key challenges and future directions for SNNs in imaging.
Main Methods:
- Structured literature survey.
- Comparative meta-analysis of datasets, training strategies, hardware, and applications.
- Case study on LMU-based depth estimation in direct Time-of-Flight (dToF) imaging.
Main Results:
- SNN progress is limited by small datasets, inefficient ANN-SNN conversion, and simulation-based evaluations.
- Accuracy-efficiency trade-offs, latency bottlenecks, and sensor-hardware integration are key issues.
- Current applications are narrowly focused on classification tasks.
Conclusions:
- Developing standardized benchmarks is crucial.
- Hardware-aware training methods are needed for improved performance.
- Expanding application domains and fostering ecosystem development are essential for SNNs in imaging.
