混合精度加速点云检测，实现实时推理且不损失精度。
Mixed Precision PointPillars for Efficient 3D Object Detection with TensorRT
- 按层逐次量化筛选敏感层，关键层保留浮点，其余转8位整数。
- 在PointPillars上实现最高2.538倍加速，精度接近原始模型。
- 仅需少量校准数据即可缓解极端值影响，适合车载部署场景。
激光雷达3D目标检测对自动驾驶至关重要，实时性要求高。模型量化可加速运行，但直接应用常因点云数据数值分布广、存在极端异常值导致性能下降。为此，本文提出适用于PointPillars的混合精度框架：通过逐层量化至8位整数（INT8）并评估平均精度（AP），识别出前k个最敏感层，将其保留为浮点（FP）；再通过贪心搜索生成候选混合精度模型，最终采用后训练量化（PTQ）或量化感知训练（QAT）优化。此外，发现使用极少量校准数据可显著降低遇到异常值的概率，提升PTQ效果。该方法在无训练的PTQ流程中生成有效混合精度模型，而QAT流程性能媲美浮点模型。在TensorRT部署下，混合精度模型相较FP32模型延迟降低最多达2.538倍。
原文摘要 · Abstract (English)
LIDAR 3D object detection is one of the important tasks for autonomous vehicles. Ensuring that this task operates in real-time is crucial. Toward this, model quantization can be used to accelerate the runtime. However, directly applying model quantization often leads to performance degradation due to LIDAR's wide numerical distributions and extreme outliers. To address the wide numerical distribution, we proposed a mixed precision framework designed for PointPillars. Our framework first searches for sensitive layers with post-training quantization (PTQ) by quantizing one layer at a time to 8-bit integer (INT8) and evaluating each model for average precision (AP). The top-k most sensitive layers are assigned as floating point (FP). Combinations of these layers are greedily searched to produce candidate mixed precision models, which are finalized with either PTQ or quantization-aware training (QAT). Furthermore, to handle outliers, we observe that using a very small number of calibration data reduces the likelihood of encountering outliers, thereby improving PTQ performance. Our methods provides mixed precision models without training in the PTQ pipeline, while our QAT pipeline achieves the performance competitive to FP models. With TensorRT deployment, our mixed precision models offer less latency by up to 2.538 times compared to FP32 models.
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。