轻量级激光雷达3D目标检测算法，专为FPGA实时推理优化。
LiFT: Lightweight, FPGA-tailored 3D object detection based on LiDAR data
- 基于FPGA限制设计，采用2D单元处理与INT8量化降低计算开销。
- 仅20.73 GMAC计算量下实现51.84% mAP和61.01% NDS性能。
- 适合边缘设备部署，尤其适用于资源受限的实时自动驾驶场景。
本文提出LiFT，一种面向FPGA平台的轻量级、全量化3D目标检测算法，专为激光雷达数据实现实时推理而设计。通过深入分析FPGA特有的限制，识别出一系列影响算法设计的约束：计算复杂度上限为30 GMAC，权重与激活值采用INT8量化，使用2D细胞而非3D体素进行处理，且尽量减少跳连结构。为在满足这些约束的同时最大化性能，LiFT结合了新型机制与前沿技术，如可重参数化卷积与全稀疏架构。核心创新包括双边界柱状特征网络，可在不增加复杂度的情况下提升性能，以及输入特征的高效INT8量化方案。算法仅需20.73 GMAC计算量，是少数针对极低复杂度3D检测的算法之一。在具有挑战性的NuScenes验证集上，其表现优于同类方法，mAP达51.84%，NDS为61.01%。代码将开源于https://github.com/vision-agh/lift。
原文摘要 · Abstract (English)
This paper presents LiFT, a lightweight, fully quantized 3D object detection algorithm for LiDAR data, optimized for real-time inference on FPGA platforms. Through an in-depth analysis of FPGA-specific limitations, we identify a set of FPGA-induced constraints that shape the algorithm's design. These include a computational complexity limit of 30 GMACs (billion multiply-accumulate operations), INT8 quantization for weights and activations, 2D cell-based processing instead of 3D voxels, and minimal use of skip connections. To meet these constraints while maximizing performance, LiFT combines novel mechanisms with state-of-the-art techniques such as reparameterizable convolutions and fully sparse architecture. Key innovations include the Dual-bound Pillar Feature Net, which boosts performance without increasing complexity, and an efficient scheme for INT8 quantization of input features. With a computational cost of just 20.73 GMACs, LiFT stands out as one of the few algorithms targeting minimal-complexity 3D object detection. Among comparable methods, LiFT ranks first, achieving an mAP of 51.84% and an NDS of 61.01% on the challenging NuScenes validation dataset. The code will be available at https://github.com/vision-agh/lift.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。