TRIM：基于激活统计的3D目标检测时序运行时监控
TRIM: Temporal Run-Time Monitoring of 3D Object Detection via Activation Statistics
Hakan Yekta Yatbaz · Konstantinos Koufos · Mehrdad Dianati
中文摘要
可靠感知对智能交通系统至关重要，但深度神经网络可能在持续输出看似合理结果的同时，静默漏检安全关键目标。现有3D目标检测运行时监控器通常逐帧独立评估，无法利用遮挡、点稀疏性和场景复杂度的时序演化。TRIM是一种轻量级时序监控器，无需修改或重新训练底层检测器即可检测漏检失败。TRIM用紧凑的通道级统计量汇总中间激活，并用时序卷积网络处理这些摘要的短序列。在NuScenes上的实验表明，与单帧监控相比，时序监控将AUROC从0.686提升至0.704，并将95%召回率下的误报率从0.818降至0.759。骨干网络早期和中间特征提供最强的监控信号，而时序推理头在CPU上仅需0.159 ms。
关键要点
- 01现有3D检测运行时监控逐帧独立评估，无法利用遮挡与点稀疏性的时序演化。
- 02TRIM用通道级统计量汇总中间激活，经时序卷积网络处理短序列，无需修改或重训检测器。
- 03NuScenes上AUROC从0.686升至0.704，95%召回率下误报率从0.818降至0.759。
- 04骨干网络早期与中间特征监控信号最强，时序推理头在CPU上仅需0.159 ms。
解读
尚无解读。
原始英文摘要
arXiv:2610.38613v1 Announce Type: cross Abstract: Reliable perception is essential to intelligent transportation systems, yet deep neural networks can silently miss safety-critical objects while continuing to produce apparently plausible outputs. Existing run-time monitors for 3D object detection generally assess each frame independently and therefore cannot exploit the temporal evolution of occlusion, point sparsity, and scene complexity. We present TRIM, a lightweight temporal monitor that detects missed-object failures without modifying or retraining the underlying detector. TRIM summarises intermediate activations using compact channel-wise statistics and processes short sequences of these summaries with a temporal convolutional network. Experiments on NuScenes show that temporal monitoring improves AUROC from 0.686 to 0.704 and reduces the false alarm rate at 95% recall from 0.818 to 0.759 compared with single-frame monitoring. Early and intermediate backbone features provide the strongest monitoring signal, while the temporal inference head requires only 0.159 ms on a CPU.