UPAQ通过剪枝与量化，让车载3D目标检测模型更省电更快。
UPAQ: A Framework for Real-Time and Energy-Efficient 3D Object Detection in Autonomous Vehicles
- 结合半结构化剪枝与量化，压缩点云与图像融合的3D检测模型。
- 在Jetson Orin Nano上实现最高5.62倍压缩、1.97倍加速、2.07倍省电。
- 适合部署在算力受限的自动驾驶车载设备，提升实时性与能效。
为提升自动驾驶车辆的感知能力，近期研究聚焦于3D目标检测器，其相比传统2D检测器提供更全面的预测，但带来更高的内存占用与计算开销。本文提出新型框架UPAQ，利用半结构化模式剪枝与量化技术，提升资源受限嵌入式自动驾驶平台上的激光雷达点云与摄像头融合3D目标检测器的效率。在Jetson Orin Nano嵌入式平台上，UPAQ相较现有最优模型压缩框架，对Pointpillar与SMOKE模型分别实现最高5.62倍和5.13倍的模型压缩率、最高1.97倍和1.86倍的推理速度提升，以及最高2.07倍和1.87倍的能耗降低。
原文摘要 · Abstract (English)
To enhance perception in autonomous vehicles (AVs), recent efforts are concentrating on 3D object detectors, which deliver more comprehensive predictions than traditional 2D object detectors, at the cost of increased memory footprint and computational resource usage. We present a novel framework called UPAQ, which leverages semi-structured pattern pruning and quantization to improve the efficiency of LiDAR point-cloud and camera-based 3D object detectors on resource-constrained embedded AV platforms. Experimental results on the Jetson Orin Nano embedded platform indicate that UPAQ achieves up to 5.62x and 5.13x model compression rates, up to 1.97x and 1.86x boost in inference speed, and up to 2.07x and 1.87x reduction in energy consumption compared to state-of-the-art model compression frameworks, on the Pointpillar and SMOKE models respectively.
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。