[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2683":3},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":6,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":18,"tags":20,"view_count":15,"doi":24,"paper":25,"created_at":42},2683,"Efficient Monocular Depth Estimation on Embedded Systems with Neural Cellular Automata","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs42979-026-05327-4","Real-time monocular depth estimation is an essential task for autonomous drone navigation, yet existing models require hundreds of billions of floating-point operations per in-ference, rendering them impractical for deployment on resource-constrained embedded systems. On the MidAir aerial imagery dataset, a mean absolute error of 5.05 m is obtained with mea-sured CPU inference latency of 3.66 ms and 40,832 parameters—2.9–6.1× faster and 5.4–16.3× fewer parameters than com-parable lightweight convolutional neural network (CNN) base-lines (FastDepth, MiniDepth, RT-MonoDepth-S, MiDaS-Lite). A depth-augmented semantic segmentation variant performs simultaneous depth estimation and semantic segmentation in a single forward pass with approximately 59,000 parameters, enabling holistic computer vision for autonomous flight. Diverse image and video processing tasks are supported by the same design, providing a practical foundation for efficient real-time computer vision in edge computing applications.","实时单目深度估计是自主无人机导航的一项关键任务，然而现有模型每次推理需要数千亿次浮点运算，使其难以部署在资源受限的嵌入式系统上。在MidAir航空影像数据集上，该方法的平均绝对误差为5.05 m，实测CPU推理延迟为3.66 ms，参数量为40,832——相比同类轻量级卷积神经网络（CNN）基线（FastDepth、MiniDepth、RT-MonoDepth-S、MiDaS-Lite），推理速度快2.9–6.1倍，参数量少5.4–16.3倍。一种深度增强语义分割变体可在单次前向传播中同时完成深度估计与语义分割，参数量约为59,000，为自主飞行提供了整体计算机视觉能力。同一设计支持多种图像与视频处理任务，为边缘计算应用中高效的实时计算机视觉提供了实用基础。",null,"SN Computer Science","2026-09-15T00:00:00Z","论文",10,false,0,{"impact":15,"substance":15,"depth":15,"authority":15,"freshness":15,"relevant":15,"comment":17},"面向嵌入式系统的单目深度估计方法论文，属通用计算机视觉与无人机导航技术，未涉及三农或农业信息化应用场景，相关性不足。",[19],{"name":10,"url":6},[21,22,23],"无人机","农业人工智能","边缘计算","10.1007\u002Fs42979-026-05327-4",{"doi":24,"openalex_id":26,"authors":27,"venue":10,"cited_by_count":15,"oa_url":33,"card":34,"direction":40,"ingested_from":41},"W7213322770",[28,30],{"name":29,"orcid":9},"Kurt Nunn",{"name":31,"orcid":32},"Tooraj Nikoubin","https:\u002F\u002Forcid.org\u002F0000-0003-1724-3503","https:\u002F\u002Flink.springer.com\u002Fcontent\u002Fpdf\u002F10.1007\u002Fs42979-026-05327-4.pdf",{"tldr":35,"method":36,"finding":37,"direction":38,"opportunity":39},"提出基于神经细胞自动机的轻量单目深度估计模型，在嵌入式系统上实现实时推理。","神经细胞自动机，MidAir航拍数据集，CPU推理，参数量仅4万。","深度估计MAE 5.05米，推理3.66毫秒，比轻量CNN快2.9-6.1倍且参数少5.4-16.3","农业遥感与作物表型","可探索该轻量模型在无人机农业场景（如作物高度估计、避障）的迁移与精度-效率平衡。","农业人工智能与决策模型","openalex","2026-09-16T23:30:52.892539Z"]