# PowerYOLO: Mixed Precision Model for Hardware Efficient Object Detection with Event Data

arXiv: https://arxiv.org/abs/2407.08272 · HTML: https://mlsanity.com/paper/2407.08272/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 32 references that are in the map.
- First posted on arXiv 11 Jul 2024.

## Builds on (most cited first, 21 of 21)

- [High Speed and High Dynamic Range Video with an Event Camera](https://mlsanity.com/paper/1906.07165/) (2019)
- [Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluation](https://mlsanity.com/paper/2004.09602/) (2020)
- [Learning to Detect Objects with a 1 Megapixel Event Camera](https://mlsanity.com/paper/2009.13436/) (2020)
- [Recurrent Vision Transformers for Object Detection with Event Cameras](https://mlsanity.com/paper/2212.05598/) (2022)
- [FINN: A Framework for Fast, Scalable Binarized Neural Network Inference](https://mlsanity.com/paper/1612.07119/) (2016)
- [A Large Scale Event-based Detection Dataset for Automotive](https://mlsanity.com/paper/2001.08499/) (2020)
- [Event-based Asynchronous Sparse Convolutional Networks](https://mlsanity.com/paper/2003.09148/) (2020)
- [Convolutional Neural Networks using Logarithmic Data Representation](https://mlsanity.com/paper/1603.01025/) (2016)
- [Deep Directly-Trained Spiking Neural Networks for Object Detection](https://mlsanity.com/paper/2307.11411/) (2023)
- [GET: Group Event Transformer for Event-Based Vision](https://mlsanity.com/paper/2310.02642/) (2023)
- [DeepShift: Towards Multiplication-Less Neural Networks](https://mlsanity.com/paper/1905.13298/) (2019)
- [Pushing the Limits of Asynchronous Graph-based Object Detection with Event Cameras](https://mlsanity.com/paper/2211.12324/) (2022)
- [Power-of-Two Quantization for Low Bitwidth and Hardware Compliant Neural Networks](https://mlsanity.com/paper/2203.05025/) (2022)
- [Motion Robust High-Speed Light-Weighted Object Detection With Event Camera](https://mlsanity.com/paper/2208.11602/) (2022)
- [Event-based Feature Extraction Using Adaptive Selection Thresholds](https://mlsanity.com/paper/1907.07853/) (2019)
- [Neural Network Compression Framework for fast model inference](https://mlsanity.com/paper/2002.08679/) (2020)
- [Block-Matching Optical Flow for Dynamic Vision Sensor- Algorithm and FPGA Implementation](https://mlsanity.com/paper/1706.05415/) (2017)
- [Reducing the Side-Effects of Oscillations in Training of Quantized YOLO Networks](https://mlsanity.com/paper/2311.05109/) (2023)
- [Automotive Object Detection via Learning Sparse Events by Spiking Neurons](https://mlsanity.com/paper/2307.12900/) (2023)
- [Energy Efficient Hardware Acceleration of Neural Networks with Power-of-Two Quantisation](https://mlsanity.com/paper/2209.15257/) (2022)
- [Q-YOLO: Efficient Inference for Real-time Object Detection](https://mlsanity.com/paper/2307.04816/) (2023)

## Built on by (most cited first, 2 of 2)

- [Adaptive Event Stream Slicing for Open-Vocabulary Event-Based Object Detection via Vision-Language Knowledge Distillation](https://mlsanity.com/paper/2510.00681/) (2025)
- [Neuromorphic Object Detection: An In-Depth Study and Future Directions](https://mlsanity.com/paper/2607.23576/) (2026)

## Abstract (arXiv)

The performance of object detection systems in automotive solutions must be as high as possible, with minimal response time and, due to the often battery-powered operation, low energy consumption. When designing such solutions, we therefore face challenges typical for embedded vision systems: the problem of fitting algorithms of high memory and computational complexity into small low-power devices. In this paper we propose PowerYOLO - a mixed precision solution, which targets three essential elements of such application. First, we propose a system based on a Dynamic Vision Sensor (DVS), a novel sensor, that offers low power requirements and operates well in conditions with variable illumination. It is these features that may make event cameras a preferential choice over frame cameras in some applications. Second, to ensure high accuracy and low memory and computational complexity, we propose to use 4-bit width Powers-of-Two (PoT) quantisation for convolution weights of the YOLO detector, with all other parameters quantised linearly. Finally, we embrace from PoT scheme and replace multiplication with bit-shifting to increase the efficiency of hardware acceleration of such solution, with a special convolution-batch normalisation fusion scheme. The use of specific sensor with PoT quantisation and special batch normalisation fusion leads to a unique system with almost 8x reduction in memory complexity and vast computational simplifications, with relation to a standard approach. This efficient system achieves high accuracy of mAP 0.301 on the GEN1 DVS dataset, marking the new state-of-the-art for such compressed model.