Nowadays, autonomous cars can drive smoothly in ordinary cases, and it is widely recognized that realistic sensor simulation will play a critical role in solving remaining corner cases by simulating them. To this end, we propose an autonomous driving simulator based upon neural radiance fields (NeRFs). Compared with existing works, ours has three notable features: (1) Instance-aware. Our simulator models the foreground instances and background environments separately with independent networks so that the static (e.g., size and appearance) and dynamic (e.g., trajectory) properties of instances can be controlled separately. (2) Modular. Our simulator allows flexible switching between different modern NeRF-related backbones, sampling strategies, input modalities, etc. We expect this modular design to boost academic progress and industrial deployment of NeRF-based autonomous driving simulation. (3) Realistic. Our simulator set new state-of-the-art photo-realism results given the best module selection. Our simulator will be open-sourced while most of our counterparts are not. Project page: https://open-air-sun.github.io/mars/.

We present Structured Neural Radiance Field (Structured-NeRF) for indoor scene representaion based on a novel hierarchical scene graph structure to organize the neural radiance field. Existing object-centric methods focus only on the inherent characteristics of objects, while overlooking the semantic and physical relationships between them. Our scene graph is adept at managing the complex real-world correlation between objects within a scene, enabling functionality beyond novel view synthesis, such as scene re-arrangement. Based on the hierarchical structure, we introduce the optimization strategy based on semantic and physical relationships, thus simplifying the operations involved in scene editing and ensuring both efficiency and accuracy. Moreover, we conduct shadow rendering on objects to further intensify the realism of the rendered images. Experimental results demonstrate our structured representation not only achieves state-of-the-art (SOTA) performance in object-level and scene-level rendering, but also advances downstream applications in union with LLM/VLM, such as automatic and instruction/image conditioned scene re-arrangement, thereby extending the NeRF to interactive editing conveniently and controllably.

为了有效确保输电线路安全性,增强无线通信功能,设计了基于无人机影像的输电线路自动巡检系统.搭建无人机巡检环境,采用无线通信模块实现无人机与地面站连接功能,利用导航模块调整无人机飞行路线和角度,使用巡检任务管理模块生成巡检任务,基于单目视觉算法获取无人机与输电线路之间安全距离.通过接收输电线路巡检影像,利用卷积神经网络模型识别输电线路巡检影像内存在的输电线路缺陷,引入ELU非线性激活函数实现输电线路自动巡检.实验结果表明,该系统具备较强的无线通信功能,且对焦耗时较短、运行稳定性较好,可有效确保输电线路安全性.

Reliable and automated 3-dimensional (3D) plant shoot segmentation is a core prerequisite for the extraction of plant phenotypic traits at the organ level. Combining deep learning and point clouds can provide effective ways to address the challenge. However, fully supervised deep learning methods require datasets to be point-wise annotated, which is extremely expensive and time-consuming. In our work, we proposed a novel weakly supervised framework, Eff-3DPSeg, for 3D plant shoot segmentation. First, high-resolution point clouds of soybean were reconstructed using a low-cost photogrammetry system, and the Meshlab-based Plant Annotator was developed for plant point cloud annotation. Second, a weakly supervised deep learning method was proposed for plant organ segmentation. The method contained (a) pretraining a self-supervised network using Viewpoint Bottleneck loss to learn meaningful intrinsic structure representation from the raw point clouds and (b) fine-tuning the pretrained model with about only 0.5% points being annotated to implement plant organ segmentation. After, 3 phenotypic traits (stem diameter, leaf width, and leaf length) were extracted. To test the generality of the proposed method, the public dataset Pheno4D was included in this study. Experimental results showed that the weakly supervised network obtained similar segmentation performance compared with the fully supervised setting. Our method achieved 95.1%, 96.6%, 95.8%, and 92.2% in the precision, recall, F1 score, and mIoU for stem–leaf segmentation for the soybean dataset and 53%, 62.8%, and 70.3% in the AP, AP@25, and AP@50 for leaf instance segmentation for the Pheno4D dataset. This study provides an effective way for characterizing 3D plant architecture, which will become useful for plant breeders to enhance selection processes. The trained networks are available at https://github.com/jieyi-one/EFF-3DPSEG .

We present ASSIST, an object-wise neural radiance field as a panoptic representation for compositional and realistic simulation. Central to our approach is a novel scene node data structure that stores the information of each object in a unified fashion, allowing online interaction in both intra- and cross-scene settings. By incorporating a differentiable neural network along with the associated bounding box and semantic features, the proposed structure guarantees user-friendly interaction on independent objects to scale up novel view simulation. Objects in the scene can be queried, added, duplicated, deleted, transformed, or swapped simply through mouse/keyboard controls or language instructions. Experiments demonstrate the efficacy of the proposed method, where scaled realistic simulation can be achieved through interactive editing and compositional rendering, with color images, depth images, and panoptic segmentation masks generated in a 3D consistent manner.

Recently, 3D scenes parsing with deep learning approaches has been a heating topic. However, current methods with fully-supervised models require manually annotated point-wise supervision which is extremely user-unfriendly and time-consuming to obtain. As such, training 3D scene parsing models with sparse supervision is an intriguing alternative. We term this task as data-efficient 3D scene parsing and propose an effective two-stage framework named VIBUS to resolve it by exploiting the enormous unlabeled points. In the first stage, we perform self-supervised representation learning on unlabeled points with the proposed Viewpoint Bottleneck loss function. The loss function is derived from an information bottleneck objective imposed on scenes under different viewpoints, making the process of representation learning free of degradation and sampling. In the second stage, pseudo labels are harvested from the sparse labels based on uncertainty-spectrum modeling. By combining data-driven uncertainty measures and 3D mesh spectrum measures (derived from normal directions and geodesic distances), a robust local affinity metric is obtained. Finite gamma/beta mixture models are used to decompose category-wise distributions of these measures, leading to automatic selection of thresholds. We evaluate VIBUS on the public benchmark ScanNet and achieve state-of-the-art results on both validation set and online test server. Ablation studies show that both Viewpoint Bottleneck and uncertainty-spectrum modeling bring significant improvements. Codes and models are publicly available at https://github.com/AIR-DISCOVER/VIBUS.

特征提取作为处理输电线路航拍巡检图像最重要阶段之一,传统方法对故障或目标的识别准确率不高且单一,耗时较长,且易受到背景、形态及材料等因素的影响,难以应用于实际中.为解决上述传统方法缺陷,引入深度学习,采用一种基于区域的全卷积网络,利用标注数据训练其网络,并利用在线困难样本挖掘、样本优化、软性非极大值抑制等改进方法进行优化.实验结果表明,所提方面在定位目标更快更准确,应用在输电线路巡检的检测精度较高,以满足输电线路智能巡检的需求.

This paper presents an intelligent machine which can automatically convert the captured portrait into a physical gadget made up of LEGO bricks. On the contrary to synthesising a 2D image or a virtual 3D object, generating physical 3D assembly object needs to take physical properties and assembly process into consideration, leading to more challenges. To generate brick models for arbitrary portraits, we formulate the transformation between the attribute space (extracted from 2D images) and the brick model space as a constraint integer programming problem which can be solved with a heuristic search method. Furthermore, as the bricks are physically scattered, we propose an algorithm to generate corresponding assembly instructions for customized figure-featured-bricks to facilitate users' assembly. Meanwhile, we deploy the proposed algorithms on an automatic machine which integrates a camera, a printer, a laptop, and a brick operation unit. Finally, the generated brick models and assembly instructions are evaluated by a large number of users. It is worth noting that the whole system works as an intelligent vending machine, producing a 150-brick-model within 3 minutes.

Three-dimensional (3D) high throughput plant phenotyping techniques provide an opportunity to measure and predict plant geometric traits and their responses to changing environmental factors, which is highly useful to speed up the selection of new genotypes under specific growing conditions. 3D plant shoot segmentation is essential to obtain plant phenotypic traits at the organ level. Convolutional neural networks have been used for plant point cloud segmentation. However, the network training needs point-wise plant annotation which is extremely expensive and time-consuming. In this work, we aimed to address the challenge of network training using a small subset of annotations, about only 0.5% of labeled points. To this end, we proposed a framework, Eff-PlantNet, which contains two stages. In the first stage, meaningful representations from the plant point clouds were learned through a 3D self-supervised representation learning network without the usage of annotation. In the second stage, a subsequent weakly-supervised fine-tuning by the pre-trained model was used to conduct point cloud segmentation. Our framework was low-cost and effective and produced similar plant semantic and instance segmentation performance compared with the full-supervised training network.

净空区异物是导致航空事故的主要原因之一,无人机技术与异物探测的结合提高了异物检测的效率.为了优化无人机自主巡检航迹,本文提出了面向机场净空区异物探测的无人机自主巡检航迹优化方法.首先,规划无人机自主巡检航水平轨迹,计算出无人机自主巡检航迹的最优长度;其次,基于最短时间算法规划航迹路径,使用冗余拍摄航点剔除实现探测的航点处理;再次,将异物探测航迹平滑处理,计算出最短的无人机自主巡检时间,实现自主巡检航迹优化;最后,通过对比实验证明优化后的自主巡检方式的各项指标优于现有方法.

In the three-dimensional scene,to improve the management efficiency of substation equipment and the accuracy of data display,this paper uses GIS technology to realize the management and data display of substation equipment.First,the laser scanner is used to obtain the information of substation equipment,and the specific location of the equipment is obtained according to GIS technology.Combined with the two kinds of information obtained above,the three-dimensional model of substation equipment is constructed.Then,the substation equipment data management database is established,and the spatial database is constructed by AreGIS Server software,which is imported into the substation equipment data management database to carry out comprehensive management.Based on this,in the graphic data display unit,the composite object technology of 2D GIS and 3D GIS is used to display the graphic data of 3D model.Experimental results show that the model has a higher efficiency of substation equipment management,can ensure a higher success rate of data display,and its imaging definition is higher.

Semantic understanding of 3D point clouds is important for various robotics applications. Given that point-wise semantic annotation is expensive, in this paper, we address the challenge of learning models with extremely sparse labels. The core problem is how to leverage numerous unlabeled points. To this end, we propose a self-supervised 3D representation learning framework named viewpoint bottleneck. It optimizes a mutual-information based objective, which is applied on point clouds under different viewpoints. A principled analysis shows that viewpoint bottleneck leads to an elegant surrogate loss function that is suitable for large-scale point cloud data. Compared with former arts based upon contrastive learning, viewpoint bottleneck operates on the feature dimension instead of the sample dimension. This paradigm shift has several advantages: It is easy to implement and tune, does not need negative samples and performs better on our goal down-streaming task. We evaluate our method on the public benchmark ScanNet, under the pointly-supervised setting. We achieve the best quantitative results among comparable solutions. Meanwhile we provide an extensive qualitative inspection on various challenging scenes. They demonstrate that our models can produce fairly good scene parsing results for robotics applications. Our code, data and models will be made public.

Murine models have been widely used to investigate the mechanobiology of aortic atherosclerosis and dissections, which develop preferably at different anatomic locations of aorta. Based MRI and finite element analysis with fluid-structure interaction, we numerically investigated factors that may affect the blood flow and structural mechanics of rat aorta. The results indicated that aortic root motion greatly increases time-averaged wall shear stress (TAWSS), oscillatory shear index (OSI), relative residence time (RRT), displacement of the aorta, and enhances helical flow pattern but has limited influence on effective stress, which is highly modulated by blood pressure. Moreover, the influence of the motion component on these indicators is different with axial motion more obvious than planar motion. Surrounding fixation of the intercostal arteries and the branch vessels on aortic arch would reduce the influence of aortic root motion. The compliance of the aorta has different influences at different regions, leading to decrease in TAWSS and helical flow, increase in OSI, RRT at the aortic arch, but has reversed effects on the branch vessels. When compared with the steady flow, the pulsatile blood flow would obviously increase the WSS, the displacement, and the effective stress in most regions. In conclusion, to accurately quantify the blood flow and structural mechanics of rat aorta, the motion of the aortic root, the compliance of aortic wall, and the pulsation of blood flow should be considered. However, when only focusing on the effective stress in rat aorta, the motion of the aortic root may be neglected.

Abstract. In haze, the quality of the images is degraded due to the scattering of atmospheric aerosol particles in near-earth remote sensing. Therefore, how to effectively remove the influence of haze and improve image quality has been a hot issue. A polarization spectral image dehazing method is proposed here. A multi-spectral full-polarization imager was used to detect the polarization spectral images of ground objects. Firstly, the maximum- and minimum-intensity polarization images were obtained from the relationship between the Stokes vector and the Mueller matrix. Secondly, the airlight polarization model was utilized to estimate the airlight radiance at an infinite distance and the degree of polarization of the airlight. At last, the atmospheric attenuation model was used to obtain dehazed images. The results proved that our proposed image dehazing method can achieve substantial improvements on the detail recovery, not only in vertical downward detection mode, but also in horizontal detection mode.

In this paper a method to estimate surface roughness of sand land from multi-angle and multi-waveband polarized detections is presented. Firstly, the polarized bidirectional reflectance distribution function (pBRDF) of the sand land's surface based on the microfacet theory was established. Then three sand samples with particle sizes of 0.5 mm, 0.7 mm and 1 mm were obtained by a series of sieves. And the polarization information was acquired by full-polarized multi-spectral imaging system based on Liquid Crystal Variable Retarder (LCVR). We used the nonlinear least squares method to estimate the surface roughness of from the measured data. Lastly, the analysis results show that the accuracy of sand roughness estimation is improved as the number of the angles (i.e., source incident angles and detection angles) and wavebands increase until the estimation accuracy saturates. It is indicated that the method based on polarization imaging detection to estimate sandy land surface roughness is effective.

Polarimetric imagery is a powerful tool in remote sensing because the polarization information of the targets contains surface features, shape, material composition, and surface roughness, which improves applications such as target detection, shape extraction and anomaly detection. For the importance of the quantitative polarmetric remote sensing, the quantitative estimation of the physical characteristics of the target has attracted considerable scientific interests and become the trend of polarimetric imagery. However, because of the spatial scale effect, specifically, in the large detection distance, the quantitative estimation accuracy of the target can be affected by the inhomogeneous target. In this paper, a novel method based on polarization imaging detection to estimate the roughness of inhomogeneous paint surface in outdoor is proposed. A shadowing method was used to eliminate the effect of skylight and improve the estimation accuracy in outdoor experiment. In addition, the correction method based on local variance of roughness distribution was performed to improve the estimation accuracy of inhomogeneous paint. The results showed that the estimation error of roughness for homogeneous paint in two distances were both below 8%. Especially for the target with smaller roughness, after the correction, the estimation accuracy of inhomogeneous paint were below 6% in two detection distances, which confirms the effectiveness of estimation approach and verifies the practicality of the correction method to improve the estimation accuracy of inhomogeneous target in polarimetric imaging remote sensing detection. The approach presented in this paper has important significance for the development of the quantitative remote sensing, especially for targets with inhomogeneous surface.

现代企业人才管理已逐渐从对"人"的管理向对"能力"的管理转变,将组织内分散在员工个体的知识,整合优化为组织整体的能力.人才管理必须紧紧围绕组织能力下功夫,维护、巩固和发展企业人才发展战略,通过后备人才"能力建模、能力摸底、能力提升、能力评估、能力规划"工作,实现对不同类别、层级后备人才的能力和组织能力的闭环管理,达到人才能力提升的根本目的.

电网企业作为技术密集型企业,专业技术人才是企业持续发展的动力和助力,打造优秀的专业技术人才队伍是企业发展战略实现的保证.为专业技术人员提供规划合理、持续发展的成长环境,是企业的可持续发展的坚实基础.而“学习地图”则是企业开展员工能力管理、员工规划职业生涯的有效工具,是企业打造专业技术人才队伍的重要基础.

在我国经济不断发展的同时，个大规模的电力企业在相应标榜业绩，而业绩的好坏与电力企业人才的引进方式以及人才的管理有着密切的关系，所以各大电力企业为了迎合经济发展的需求，对人才管理格外重视。人才管理对电力企业有着不可言喻的重要作用，对电力企业生产效益及社会效益均有提高，平衡电力企业负面状态。本文就是透过电力企业人力资源管理对党政人才管理进行完善，从而从各个角度强化对人民利益的回馈。