通过动态识别感兴趣区域，降低自动驾驶感知的内存能耗。
MotiMem: Motion-Aware Approximate Memory for Energy-Efficient Neural Perception in Autonomous Vehicles
- 利用运动传播动态定位图像关键区域，减少无效数据传输。
- 结合自适应截断与反转，实现比特级稀疏编码，能耗降43%。
- 适合电池受限的自动驾驶系统，兼顾能效与检测精度。
高分辨率传感器对鲁棒自动驾驶感知至关重要，但给电池受限的电动车带来严重内存瓶颈。此类系统中，数据移动能耗远超计算能耗。传统图像压缩因语义无感知且针对存储优化，不适用于此场景。本文提出MotiMem，一种软硬件协同设计的接口。利用时间一致性，MotiMem采用轻量级二维运动传播动态识别感兴趣区域（RoI）。同时，混合稀疏感知编码方案通过自适应反演与截断诱导比特级稀疏性。在nuScenes、Waymo和KITTI上，使用16种检测模型的大量实验表明，MotiMem将内存接口动态能耗降低约43%，同时保持约93%的目标检测精度，显著优于JPEG、WebP等标准编解码器，确立了新的能效-精度权衡前沿。
原文摘要 · Abstract (English)
High-resolution sensors are critical for robust autonomous perception but impose a severe memory wall on battery-constrained electric vehicles. In these systems, data movement energy often outweighs computation. Traditional image compression is ill-suited as it is semantically blind and optimizes for storage rather than bus switching activity. We propose MotiMem, a hardware-software co-designed interface. Exploiting temporal coherence,MotiMem uses lightweight 2D Motion Propagation to dynamically identify Regions of Interest (RoI). Complementing this, a Hybrid Sparsity-Aware Coding scheme leverages adaptive inversion and truncation to induce bitlevel sparsity. Extensive experiments across nuScenes, Waymo, and KITTI with 16 detection models demonstrate that MotiMem reduces memory-interface dynamic energy by approximately 43 percent while retaining approximately 93 percent of the object detection accuracy, establishing a new Pareto frontier significantly superior to standard codecs like JPEG and WebP.
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。