轻量级单目视觉系统实现嵌入式微出行路径精准提取
Lightweight Machine Learning-Driven Monocular Sidewalk Path Extraction for Embedded Micromobility Navigation
- 采用轻量图像空间架构与师生学习框架提升分割效率
- 图像空间规划速度比鸟瞰图快421倍，误差仅14.3像素
- 适合低功耗嵌入式设备，实测端到端延迟低于50毫秒
在杂乱且地图稀疏的环境中，微型出行系统需要在紧凑低功耗硬件上可靠完成感知与路径规划。本文提出一种单目视觉路径提取流水线，历经三轮迭代，从骨架图基线到距离变换走廊规划，最终形成轻量图像空间架构，并系统比较了五种路径规划方法在鸟瞰图（BEV）与图像空间中的表现。一个基于半监督师生框架训练的紧凑SegFormer-B0学生模型，在使用OneFormer Swin-L伪标签时，达到11.7毫秒/帧的推理速度和0.946的标注交并比（IoU），优于基线（IoU 0.758，18.9毫秒）。在32帧人工标注数据上的对比显示，图像空间中点规划横向中心误差仅为14.3像素，速度达2.2毫秒，相比鸟瞰图距离变换规划（926.8毫秒，65.0像素误差）提速421倍，同时保持相近的掩码-路径对齐率（98.5% vs 98.6%）。六段校园视频全帧回放（共22,679帧）验证：改进分割使时间不稳定性从1.46%降至0.33%，模板路径可用性从73.7%升至79.3%。进一步表明，纯鸟瞰图路径提取在单目场景下脆弱——一次测试中99.3%帧无法生成有效路径。最终推荐架构为：以图像空间中点为主、图像空间距离变换为备选、鸟瞰图仅用于可视化，全链路端到端处理可在CPU上保持每帧低于50毫秒，适用于嵌入式行人级微出行系统。
原文摘要 · Abstract (English)
Sidewalk-scale path extraction demands perception and planning that run reliably on compact, low-power hardware in cluttered, map-sparse environments. We present a monocular vision pipeline for sidewalk path extraction in micromobility systems that progresses through three design iterations, from a skeleton-graph baseline through distance-transform corridor planning to a lightweight image-space architecture, and provides a systematic comparison of five path-planning methods across both bird's-eye-view (BEV) and image-space domains. A compact SegFormer-B0 student model, trained with a semi-supervised teacher-student framework using OneFormer Swin-L pseudo-labels, achieves a hand-annotated IoU of 0.946 at 11.7 ms per frame, improving over the baseline checkpoint (IoU 0.758, 18.9 ms). In a controlled planner comparison on 32 hand-labeled frames, image-space midpoint planning achieves the lowest lateral center error (14.3 px) at 2.2 ms, a 421x speedup over BEV distance-transform planning (926.8 ms, 65.0 px center error), while maintaining comparable mask-path alignment (98.5% versus 98.6%). A full-video replay across six campus sequences (22,679 frames) confirms that the improved segmentation reduces temporal instability from 1.46% to 0.33% and increases template-path availability from 73.7% to 79.3%. We further show that BEV-only path extraction is fragile in monocular settings: in one profiled run, 99.3% of frames produced no valid BEV path. The final recommended architecture, image-space midpoint primary, image-space distance-transform fallback, and BEV reserved for visualization, runs the full perception-to-path stack in under 50 ms per frame on CPU, making it suitable for embedded pedestrian-speed micromobility systems.
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。