GLANCE: Gaze-Led Attention Network for Compressed Edge-inference

TL;DR

This paper introduces GLANCE, a gaze-led attention network that enables real-time, energy-efficient object detection for AR/VR devices by combining gaze estimation with region-focused detection, achieving high accuracy with low latency.

Contribution

The paper presents a novel memory-centric, attention-guided detection pipeline that significantly reduces computation and energy use while maintaining high detection accuracy on resource-constrained devices.

Findings

Achieves 48.1% mAP on COCO with sub-10ms latency.

Reduces computational load by 40-50% and energy consumption by 65%.

Outperforms baseline YOLOv12n in accuracy on small, medium, and large objects.

Abstract

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.