# Realtime Facial Expression Recognition: Neuromorphic Hardware vs. Edge AI Accelerators

arXiv: https://arxiv.org/abs/2403.08792 · HTML: https://mlsanity.com/paper/2403.08792/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 23 references that are in the map.
- First posted on arXiv 30 Jan 2024.

## Builds on (most cited first, 7 of 7)

- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Spiking Neural Networks and Online Learning: An Overview and Perspectives](https://mlsanity.com/paper/1908.08019/) (2019)
- [Accelerator-aware Neural Network Design using AutoML](https://mlsanity.com/paper/2003.02838/) (2020)
- [Static Hand Gesture Recognition for American Sign Language using Neuromorphic Hardware](https://mlsanity.com/paper/2207.12559/) (2022)
- [Facial Expression Recognition at the Edge: CPU vs GPU vs VPU vs TPU](https://mlsanity.com/paper/2305.15422/) (2023)
- [An Adaptive Sampling and Edge Detection Approach for Encoding Static Images for Spiking Neural Networks](https://mlsanity.com/paper/2110.10217/) (2021)
- [Energy-Efficient Deployment of Machine Learning Workloads on Neuromorphic Hardware](https://mlsanity.com/paper/2210.05006/) (2022)

## Built on by (most cited first, 4 of 4)

- [LLMPi: Optimizing LLMs for High-Throughput on Raspberry Pi](https://mlsanity.com/paper/2504.02118/) (2025)
- [Towards Efficient Deployment of Hybrid SNNs on Neuromorphic and Edge AI Hardware](https://mlsanity.com/paper/2407.08704/) (2024)
- [Flex-TPU: A Flexible TPU with Runtime Reconfigurable Dataflow Architecture](https://mlsanity.com/paper/2407.08700/) (2024)
- [Threat Modeling for Enhancing Security of IoT Audio Classification Devices under a Secure Protocols Framework](https://mlsanity.com/paper/2509.14657/) (2025)

## Abstract (arXiv)

The paper focuses on real-time facial expression recognition (FER) systems as an important component in various real-world applications such as social robotics. We investigate two hardware options for the deployment of FER machine learning (ML) models at the edge: neuromorphic hardware versus edge AI accelerators. Our study includes exhaustive experiments providing comparative analyses between the Intel Loihi neuromorphic processor and four distinct edge platforms: Raspberry Pi-4, Intel Neural Compute Stick (NSC), Jetson Nano, and Coral TPU. The results obtained show that Loihi can achieve approximately two orders of magnitude reduction in power dissipation and one order of magnitude energy savings compared to Coral TPU which happens to be the least power-intensive and energy-consuming edge AI accelerator. These reductions in power and energy are achieved while the neuromorphic solution maintains a comparable level of accuracy with the edge accelerators, all within the real-time latency requirements.