EdgeLPR: On the Deep Neural Network trade-off between Precision and Performance in LiDAR Place Recognition
For practitioners deploying place recognition on resource-constrained platforms, this work provides a benchmark of quantization trade-offs, though the findings are incremental.
This paper explores efficient LiDAR-based place recognition for EdgeAI using Bird's Eye View representations and lightweight image-based networks, benchmarking architectures under FP32, FP16, and INT8 quantization. Results show FP16 matches FP32 with lower cost, while INT8 introduces architecture-dependent degradation.
Place recognition is essential for long-term autonomous navigation, enabling loop closure and consistent mapping. Although deep learning has improved performance, deploying such models on resource-constrained platforms remains challenging. This work explores efficient LiDAR-based place recognition for EdgeAI by leveraging Bird's Eye View representations to enable lightweight image-based networks. We benchmark representative architectures without aggregation heads using a unified descriptor scheme based on global pooling and linear projection, and evaluate performance under FP32, FP16, and INT8 quantization. Experiments reveal trade-offs between accuracy, robustness, and efficiency: FP16 matches FP32 with lower cost, while INT8 introduces architecture-dependent degradation. Overall, the presented results are a strong basis for future research on 'use-case'-aware quantisation of Neural Networks for Edge deployment.