用眼球追踪引导检测，让可穿戴设备实时高效识别目标。
GLANCE: Gaze-Led Attention Network for Compressed Edge-inference
- 用查表替代计算实现超低功耗眼球追踪
- 检测区域聚焦注意力，降低40%-50%计算量
- 适合资源受限的AR/VR可穿戴设备
AR/VR系统中的实时目标检测面临严峻的计算约束，要求在极低功耗下实现亚10毫秒延迟。受生物视网膜中央凹视觉启发，我们提出一种两阶段流程：通过可微无权神经网络实现超高效眼球追踪，再以注意力引导感兴趣区域的目标检测。该方法通过内存查找替代乘加运算，实现每帧仅393次乘加操作（MACs）和2.2 KiB内存开销，角度误差为8.32°。眼球预测引导关注区域检测，使计算负担减少40%-50%，能耗降低65%。部署于Arduino Nano 33 BLE平台，在COCO数据集上取得48.1% mAP（关注物体达51.8%），延迟低于10毫秒，通信时间提升177倍。相比全局YOLOv12n基线（小/中/大目标准确率分别为39.2%、63.4%、83.1%），基于兴趣区域的方法分别提升至51.3%、72.1%、88.1%。结果表明，显式注意力建模的内存中心架构在资源受限可穿戴设备上优于均匀处理。
原文摘要 · Abstract (English)
Real-time object detection in AR/VR systems faces critical computational constraints, requiring sub-10\,ms latency within tight power budgets. Inspired by biological foveal vision, we propose a two-stage pipeline that combines differentiable weightless neural networks for ultra-efficient gaze estimation with attention-guided region-of-interest object detection. Our approach eliminates arithmetic-intensive operations by performing gaze tracking through memory lookups rather than multiply-accumulate computations, achieving an angular error of $8.32^{\circ}$ with only 393 MACs and 2.2 KiB of memory per frame. Gaze predictions guide selective object detection on attended regions, reducing computational burden by 40-50\% and energy consumption by 65\%. Deployed on the Arduino Nano 33 BLE, our system achieves 48.1\% mAP on COCO (51.8\% on attended objects) while maintaining sub-10\,ms latency, meeting stringent AR/VR requirements by improving the communication time by $\times 177$. Compared to the global YOLOv12n baseline, which achieves 39.2\%, 63.4\%, and 83.1\% accuracy for small, MEDium, and LARGE objects, respectively, the ROI-based method yields 51.3\%, 72.1\%, and 88.1\% under the same settings. This work shows that memory-centric architectures with explicit attention modeling offer better efficiency and accuracy for resource-constrained wearable platforms than uniform processing.
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。