Depth-Guided Vision Transformer With Normalizing Flows for Monocular 3D Object Detection

Monocular 3D object detection is challenging due to the lack of accurate depth information.Some methods estimate the pixel-wise depth maps from off-the-shelf depth estimators and then use them as an additional input to augment the RGB images.Depth-based methods attempt to convert estimated depth maps to pseudo-LiDAR and then use LiDAR-based object detectors or focus on the perspective of image and depth fusion learning.However,they demonstrate limited performance and efficiency as a result of depth inaccuracy and complex fusion mode with convolutions.Different from these approaches,our proposed depth-guided vision transformer with a normalizing flows(NF-DVT)network uses normalizing flows to build priors in depth maps to achieve more accurate depth information.Then we develop a novel Swin-Transformer-based backbone with a fusion module to process RGB image patches and depth map patches with two separate branches and fuse them using cross-attention to exchange information with each other.Furthermore,with the help of pixel-wise relative depth values in depth maps,we develop new relative position embeddings in the cross-attention mechanism to capture more accurate sequence ordering of input tokens.Our method is the first Swin-Transformer-based backbone architecture for monocular 3D object detection.The experimental results on the KITTI and the challenging Waymo Open datasets show the effectiveness of our proposed method and superior performance over previous counterparts.

3D Vehicle Detection Algorithm Based onMultimodal Decision-Level Fusion

3D vehicle detection based on LiDAR-camera fusion is becoming an emerging research topic in autonomous driving.The algorithm based on the Camera-LiDAR object candidate fusion method(CLOCs)is currently considered to be a more effective decision-level fusion algorithm,but it does not fully utilize the extracted features of 3D and 2D.Therefore,we proposed a 3D vehicle detection algorithm based onmultimodal decision-level fusion.First,project the anchor point of the 3D detection bounding box into the 2D image,calculate the distance between 2D and 3D anchor points,and use this distance as a new fusion feature to enhance the feature redundancy of the network.Subsequently,add an attention module:squeeze-and-excitation networks,weight each feature channel to enhance the important features of the network,and suppress useless features.The experimental results show that the mean average precision of the algorithm in the KITTI dataset is 82.96%,which outperforms previous state-ofthe-art multimodal fusion-based methods,and the average accuracy in the Easy,Moderate and Hard evaluation indicators reaches 88.96%,82.60%,and 77.31%,respectively,which are higher compared to the original CLOCs model by 1.02%,2.29%,and 0.41%,respectively.Compared with the original CLOCs algorithm,our algorithm has higher accuracy and better performance in 3D vehicle detection.

MFF-Net: Multimodal Feature Fusion Network for 3D Object Detection

In complex traffic environment scenarios,it is very important for autonomous vehicles to accurately perceive the dynamic information of other vehicles around the vehicle in advance.The accuracy of 3D object detection will be affected by problems such as illumination changes,object occlusion,and object detection distance.To this purpose,we face these challenges by proposing a multimodal feature fusion network for 3D object detection(MFF-Net).In this research,this paper first uses the spatial transformation projection algorithm to map the image features into the feature space,so that the image features are in the same spatial dimension when fused with the point cloud features.Then,feature channel weighting is performed using an adaptive expression augmentation fusion network to enhance important network features,suppress useless features,and increase the directionality of the network to features.Finally,this paper increases the probability of false detection and missed detection in the non-maximum suppression algo-rithm by increasing the one-dimensional threshold.So far,this paper has constructed a complete 3D target detection network based on multimodal feature fusion.The experimental results show that the proposed achieves an average accuracy of 82.60%on the Karlsruhe Institute of Technology and Toyota Technological Institute(KITTI)dataset,outperforming previous state-of-the-art multimodal fusion networks.In Easy,Moderate,and hard evaluation indicators,the accuracy rate of this paper reaches 90.96%,81.46%,and 75.39%.This shows that the MFF-Net network has good performance in 3D object detection.

Monocular 3D object detection with Pseudo-LiDAR confidence sampling and hierarchical geometric feature extraction in 6G network

The high bandwidth and low latency of 6G network technology enable the successful application of monocular 3D object detection on vehicle platforms.Monocular 3D-object-detection-based Pseudo-LiDAR is a low-cost,lowpower solution compared to LiDAR solutions in the field of autonomous driving.However,this technique has some problems,i.e.,(1)the poor quality of generated Pseudo-LiDAR point clouds resulting from the nonlinear error distribution of monocular depth estimation and(2)the weak representation capability of point cloud features due to the neglected global geometric structure features of point clouds existing in LiDAR-based 3D detection networks.Therefore,we proposed a Pseudo-LiDAR confidence sampling strategy and a hierarchical geometric feature extraction module for monocular 3D object detection.We first designed a point cloud confidence sampling strategy based on a 3D Gaussian distribution to assign small confidence to the points with great error in depth estimation and filter them out according to the confidence.Then,we present a hierarchical geometric feature extraction module by aggregating the local neighborhood features and a dual transformer to capture the global geometric features in the point cloud.Finally,our detection framework is based on Point-Voxel-RCNN(PV-RCNN)with high-quality Pseudo-LiDAR and enriched geometric features as input.From the experimental results,our method achieves satisfactory results in monocular 3D object detection.

3D Object Detection with Attention:Shell-Based Modeling

LIDAR point cloud-based 3D object detection aims to sense the surrounding environment by anchoring objects with the Bounding Box(BBox).However,under the three-dimensional space of autonomous driving scenes,the previous object detection methods,due to the pre-processing of the original LIDAR point cloud into voxels or pillars,lose the coordinate information of the original point cloud,slow detection speed,and gain inaccurate bounding box positioning.To address the issues above,this study proposes a new two-stage network structure to extract point cloud features directly by PointNet++,which effectively preserves the original point cloud coordinate information.To improve the detection accuracy,a shell-based modeling method is proposed.It roughly determines which spherical shell the coordinates belong to.Then,the results are refined to ground truth,thereby narrowing the localization range and improving the detection accuracy.To improve the recall of 3D object detection with bounding boxes,this paper designs a self-attention module for 3D object detection with a skip connection structure.Some of these features are highlighted by weighting them on the feature dimensions.After training,it makes the feature weights that are favorable for object detection get larger.Thus,the extracted features are more adapted to the object detection task.Extensive comparison experiments and ablation experiments conducted on the KITTI dataset verify the effectiveness of our proposed method in improving recall and precision.

Point Cloud Processing Methods for 3D Point Cloud Detection Tasks

Light detection and ranging(LiDAR)sensors play a vital role in acquiring 3D point cloud data and extracting valuable information about objects for tasks such as autonomous driving,robotics,and virtual reality(VR).However,the sparse and disordered nature of the 3D point cloud poses significant challenges to feature extraction.Overcoming limitations is critical for 3D point cloud processing.3D point cloud object detection is a very challenging and crucial task,in which point cloud processing and feature extraction methods play a crucial role and have a significant impact on subsequent object detection performance.In this overview of outstanding work in object detection from the 3D point cloud,we specifically focus on summarizing methods employed in 3D point cloud processing.We introduce the way point clouds are processed in classical 3D object detection algorithms,and their improvements to solve the problems existing in point cloud processing.Different voxelization methods and point cloud sampling strategies will influence the extracted features,thereby impacting the final detection performance.

3D obstacle detection of indoor mobile robots by floor detection and rejection

Obstacle detection is essential for mobile robots to avoid collision with obstacles.Mobile robots usually operate in indoor environments,where they encounter various kinds of obstacles;however,2D range sensor can sense obstacles only in 2D plane.In contrast,by using 3D range sensor,it is possible to detect ground and aerial obstacles that 2D range sensor cannot sense.In this paper,we present a 3D obstacle detection method that will help overcome the limitations of 2D range sensor with regard to obstacle detection.The indoor environment typically consists of a flat floor.The position of the floor can be determined by estimating the plane using the least squares method.Having determined the position of the floor,the points of obstacles can be known by rejecting the points of the floor.In the experimental section,we show the results of this approach using a Kinect sensor.

Traffic Accident Detection Based on Deformable Frustum Proposal and Adaptive Space Segmentation

Road accident detection plays an important role in abnormal scene reconstruction for Intelligent Transportation Systems and abnormal events warning for autonomous driving.This paper presents a novel 3D object detector and adaptive space partitioning algorithm to infer traffic accidents quantitatively.Using 2D region proposals in an RGB image,this method generates deformable frustums based on point cloud for each 2D region proposal and then frustum-wisely extracts features based on the farthest point sampling network(FPS-Net)and feature extraction network(FE-Net).Subsequently,the encoder-decoder network(ED-Net)implements 3D-oriented bounding box(OBB)regression.Meanwhile,the adaptive least square regression(ALSR)method is proposed to split 3D OBB.Finally,the reduced OBB intersection test is carried out to detect traffic accidents via separating surface theorem(SST).In the experiments of KITTI benchmark,our proposed 3D object detector outperforms other state-of-theartmethods.Meanwhile,collision detection algorithm achieves the satisfactory performance of 91.8%accuracy on our SHTA dataset.

ERROR ANALYSIS OF 3D DETECTING SYSTEM BASED ON WHOLE-FIELD PARALLEL CONFOCAL MICROSCOPE

Compared with the traditional scanning confocal microscopy, the effect of various factors on characteristic in multi-beam parallel confocal system is discussed, the error factors in multi-beam parallel confocal system are analyzed. The factors influencing the characteristics of the multi-beam parallel confocal system are discussed. The construction and working principle of the non-scanning 3D detecting system is introduced, and some experiment results prove the effect of various factors on the detecting system.

Spatial-Temporal Correlation 3D Vehicle Detection and Tracking System with Multiple Surveillance Cameras

Compared to 3D object detection using a single camera,multiple cameras can overcome some limitations on field-of-view,occlusion,and low detection confidence.This study employs multiple surveillance cameras and develops a cooperative 3D object detection and tracking framework by incorporating temporal and spatial information.The framework consists of a 3D vehicle detection model,cooperatively spatial-temporal relation scheme,and heuristic camera constellation method.Specifically,the proposed cross-camera association scheme combines the geometric relationship between multiple cameras and objects in corresponding detections.The spatial-temporal method is designed to associate vehicles between different points of view at a single timestamp and fulfill vehicle tracking in the time aspect.The proposed framework is evaluated based on a synthetic cooperative dataset and shows high reliability,where the cooperative perception can recall more than 66%of the trajectory instead of 11%for single-point sensing.This could contribute to full-range surveillance for intelligent transportation systems.

V2I Based Environment Perception for Autonomous Vehicles at Intersections

In recent years,autonomous driving technology has made good progress,but the noncooperative intelligence of vehicle for autonomous driving still has many technical bottlenecks when facing urban road autonomous driving challenges.V2I(Vehicle-to-Infrastructure)communication is a potential solution to enable cooperative intelligence of vehicles and roads.In this paper,the RGB-PVRCNN,an environment perception framework,is proposed to improve the environmental awareness of autonomous vehicles at intersections by leveraging V2I communication technology.This framework integrates vision feature based on PVRCNN.The normal distributions transform(NDT)point cloud registration algorithm is deployed both on onboard and roadside to obtain the position of the autonomous vehicles and to build the local map objects detected by roadside multi-sensor system are sent back to autonomous vehicles to enhance the perception ability of autonomous vehicles for benefiting path planning and traffic efficiency at the intersection.The field-testing results show that our method can effectively extend the environmental perception ability and range of autonomous vehicles at the intersection and outperform the PointPillar algorithm and the VoxelRCNN algorithm in detection accuracy.

ARM3D:Attention-based relation module for indoor 3D object detection

Relation contexts have been proved to be useful for many challenging vision tasks.In the field of3D object detection,previous methods have been taking the advantage of context encoding,graph embedding,or explicit relation reasoning to extract relation contexts.However,there exist inevitably redundant relation contexts due to noisy or low-quality proposals.In fact,invalid relation contexts usually indicate underlying scene misunderstanding and ambiguity,which may,on the contrary,reduce the performance in complex scenes.Inspired by recent attention mechanism like Transformer,we propose a novel 3D attention-based relation module(ARM3D).It encompasses objectaware relation reasoning to extract pair-wise relation contexts among qualified proposals and an attention module to distribute attention weights towards different relation contexts.In this way,ARM3D can take full advantage of the useful relation contexts and filter those less relevant or even confusing contexts,which mitigates the ambiguity in detection.We have evaluated the effectiveness of ARM3D by plugging it into several state-of-the-art 3D object detectors and showing more accurate and robust detection results.Extensive experiments show the capability and generalization of ARM3D on 3D object detection.Our source code is available at https://github.com/lanlan96/ARM3D.

RGB Image‑ and Lidar‑Based 3D Object Detection Under Multiple Lighting Scenarios

In recent years,camera-and lidar-based 3D object detection has achieved great progress.However,the related researches mainly focus on normal illumination conditions;the performance of their 3D detection algorithms will decrease under low lighting scenarios such as in the night.This work attempts to improve the fusion strategies on 3D vehicle detection accuracy in multiple lighting conditions.First,distance and uncertainty information is incorporated to guide the“painting”of semantic information onto point cloud during the data preprocessing.Moreover,a multitask framework is designed,which incorpo-rates uncertainty learning to improve detection accuracy under low-illumination scenarios.In the validation on KITTI and Dark-KITTI benchmark,the proposed method increases the vehicle detection accuracy on the KITTI benchmark by 1.35%and the generality of the model is validated on the proposed Dark-KITTI dataset,with a gain of 0.64%for vehicle detection.

PointGAT: Graph attention networks for 3D object detection

3D object detection is a critical technology in many applications,and among the various detection methods,pointcloud-based methods have been the most popular research topic in recent years.Since Graph Neural Network(GNN)is considered to be effective in dealing with pointclouds,in this work,we combined it with the attention mechanism and proposed a 3D object detection method named PointGAT.Our proposed PointGAT outperforms previous approaches on the KITTI test dataset.Experiments in real campus scenarios also demonstrate the potential of our method for further applications.

LWD-3D:Lightweight Detector Based on Self-Attention for 3D Object Detection

Lightweight modules play a key role in 3D object detection tasks for autonomous driving,which are necessary for the application of 3D object detectors.At present,research still focuses on constructing complex models and calculations to improve the detection precision at the expense of the running rate.However,building a lightweight model to learn the global features from point cloud data for 3D object detection is a significant problem.In this paper,we focus on combining convolutional neural networks with selfattention-based vision transformers to realize lightweight and high-speed computing for 3D object detection.We propose lightweight detection 3D(LWD-3D),which is a point cloud conversion and lightweight vision transformer for autonomous driving.LWD-3D utilizes a one-shot regression framework in 2D space and generates a 3D object bounding box from point cloud data,which provides a new feature representation method based on a vision transformer for 3D detection applications.The results of experiment on the KITTI 3D dataset show that LWD-3D achieves real-time detection(time per image<20 ms).LWD-3D obtains a mean average precision(mAP)75%higher than that of another 3D real-time detector with half the number of parameters.Our research extends the application of visual transformers to 3D object detection tasks.

基于遮挡率优化算法的火气系统布点设计与评估技术

本文简述了FGS火气系统布点设计与评估软件火焰检测不准确的问题,探讨了火焰探测器在不同遮挡模式下检测有效性的差异。通过研究红外辐射能量与遮挡模式、遮挡率的关系,引入火焰遮挡模型优化算法,并应用于专业的火气系统布点设计与评估软件Detect3D。该软件采用定量的方法,计算火焰、可燃气体、有毒气体等探测器的覆盖率,对火气探测器进行布点设计,以及评估和优化,以提升探测器覆盖有效性,是一种减缓风险及危险事故后果严重性的有效技术手段。

A volumetric change detection framework using UAV oblique photogrammetry–a case study of ultra-high-resolution monitoring of progressive building collapse

In this paper,we present a case study that performs an unmanned aerial vehicle(UAV)based fine-scale 3D change detection and monitoring of progressive collapse performance of a building during a demolition event.Multi-temporal oblique photogrammetry images are collected with 3D point clouds generated at different stages of the demolition.The geometric accuracy of the generated point clouds has been evaluated against both airborne and terrestrial LiDAR point clouds,achieving an average distance of 12 cm and 16 cm for roof and façade respectively.We propose a hierarchical volumetric change detection framework that unifies multi-temporal UAV images for pose estimation(free of ground control points),reconstruction,and a coarse-to-fine 3D density change analysis.This work has provided a solution capable of addressing change detection on full 3D time-series datasets where dramatic scene content changes are presented progressively.Our change detection results on the building demolition event have been evaluated against the manually marked ground-truth changes and have achieved an F-1 score varying from 0.78 to 0.92,with consistently high precision(0.92–0.99).Volumetric changes through the demolition progress are derived from change detection and have been shown to favorably reflect the qualitative and quantitative building demolition progression.

Rapid development methodology of agricultural robot navigation system working in GNSS-denied environment

Robotic autonomous operating systems in global n40avigation satellite system(GNSS)-denied agricultural environments(green houses,feeding farms,and under canopy)have recently become a research hotspot.3D light detection and ranging(LiDAR)locates the robot depending on environment and has become a popular perception sensor to navigate agricultural robots.A rapid development methodology of a 3D LiDAR-based navigation system for agricultural robots is proposed in this study,which includes:(i)individual plant clustering and its location estimation method(improved Euclidean clustering algorithm);(ii)robot path planning and tracking control method(Lyapunov direct method);(iii)construction of a robot-LiDAR-plant unified virtual simulation environment(combination use of Gazebo and SolidWorks);and(vi)evaluating the accuracy of the navigation system(triple evaluation:virtual simulation test,physical simulation test,and field test).Applying the proposed methodology,a navigation system for a grape field operation robot has been developed.The virtual simulation test,physical simulation test with GNSS as ground truth,and field test with path tracer showed that the robot could travel along the planned path quickly and smoothly.The maximum and mean absolute errors of path tracking are 2.72 cm,1.02 cm;3.12 cm,1.31 cm,respectively,which meet the accuracy requirements of field operations,establishing the effectiveness of the proposed methodology.The proposed methodology has good scalability and can be implemented in a wide variety of field robot,which is promising to shorten the development cycle of agricultural robot navigation system working in GNSS-denied environment.

Fabrication of the ZnO/NiO p–n junction foam for the enhanced sensing performance

P-Type NiO foam with rough nanostructured surface was prepared by the surface treatment of Ni foam,and then it was decorated with n-type ZnO nanopyramids to construct a 3D p–n junction foam. The p–n junction foam was used for electrochemical detection of dopamine and the sensing performance was improved significantly compared with the single NiO and ZnO. High sensitivity（171 mμA/mmol/L）, fast response（2 s）, excellent selectivity and stability were achieved. It was attributed to the introduction of numerous p–n junction interfaces, the interfacial potential barrier played as a tuning factor for the electrochemical determination of dopamine. The results demonstrated it would be an important way to improve the biosensing performance by introducing the p–n junction interfaces.