Pillar-Based 3D Object Detection from Point Cloud with Multiattention Mechanism

Traditionally, point cloud-based 3D object detectors are trained on annotated, non-sequential samples taken from driving sequences (e.g. the KITTI dataset). However, by doing this, the developed algorithms renounce to exploit any dynamic information from the driving sequences. It is reasonable to think that this information, which is available at test time when deploying the models in the experimental vehicles, could have significant predictive potential for the object detection task. To study the advantages that this kind of information could provide, we construct a dataset of dynamic occupancy grid maps from the raw KITTI dataset and find the correspondence to each of the KITTI 3D object detection dataset samples. By training a Lidar-based state-of-the-art 3D object detector with and without the dynamic information we get insights into the predictive value of the dynamics. Our results show that having access to the environment dynamics improves by 27% the ability of the detection algorithm to predict the orientation of smaller obstacles such as pedestrians. Furthermore, the 3D and bird's eye view bounding box predictions for pedestrians in challenging cases also see a 7% improvement. Qualitatively speaking, the dynamics help with the detection of partially occluded and far-away obstacles. We illustrate this fact with numerous qualitative prediction results.

Abstract3D object detection using point cloud is an essential task for autonomous driving. With the development of infrastructures, roadside perception can extend the view range of the autonomous vehicles through communication technology. Computation time and power consumption are two main concerns when deploying object detection tasks, and a light-weighted detection model applied in an embedded system is a convenient solution for both roadside and vehicleside. In this study, a 3D Point cLoud Object deTection (PLOT) network is proposed to reduce heavy computing and ensure real-time object detection performance in an embedded system. First, a bird’s eye view representation of the point cloud is calculated using pillar-based encoding method. Then a cross-stage partial network-based backbone and a feature pyramid network-based neck are implemented to generate the high-dimensional feature map. Finally, a multioutput head using a shared convolutional layer is attached to predict classes, bounding boxes, and the orientations of the objects at the same time. Extensive experiments using the Waymo Open Dataset and our own dataset are conducted to demonstrate the accuracy and efficiency of the proposed method.

Light Detection And Ranging (LiDAR) sensors can generate a number of sequential 3D point clouds, which are widely deployed in many real-world systems. 3D object detection in point clouds, is one of the most fundamental tasks. Unfortunately, the existing 3D object detection methods degrade in snowy weather, because in that situation the annotated samples are difficult to collect. To solve this issue, we propose a novel GAN-based Snowfall Point-cloud AugmentOR (GAN spaor ) to generate high-quality synthetic snowfall point clouds as augmentations. The basic idea of GAN spaor is to transfer annotated point clouds to snowfall versions by simultaneously learning the global style of real snowfall point clouds and the local details of physics-induced ones. Our framework fuses data-driven and physical modeling methods for rapidly generating data in snowy weather. To evaluate the effectiveness of GAN spaor , we employ a number of recent 3D object detection methods and train them by using the synthetic samples of GAN spaor as auxiliary augmentations. Moreover, we conduct a comparative analysis of the characteristics of the data distributions of the snowy point clouds synthesized by GAN spaor . Experimental results demonstrate that GAN spaor can improve the performance of 3D object detection methods compared with other existing snowfall point cloud simulators.

In response to the problem that the detection precision of the current 3D object detection algorithm is low when the object is severely occluded, this study proposes an object detection algorithm based on the reconstruction of sparse point clouds in the viewing frustum. The algorithm obtains more local feature information of the sparse point clouds in the viewing frustum through dimensional expansion, performs the fusion of local and global feature information of the point cloud data to obtain point cloud data with more complete semantic information, and then applies the obtained data to the 3D object detection task. The experimental results show that the precision of object detection in both 3D view and BEV (Bird’s Eye View) can be improved effectively through the algorithm, especially object detection of moderate and hard levels when the object is severely occluded. In the 3D view, the average precision of the 3D detection of cars, pedestrians, and cyclists at a moderate level can be increased by 7.1p.p., 16.39p.p., and 5.42p.p., respectively; in BEV, the average precision of the 3D detection of car, pedestrians, and cyclists at hard level can be increased by 6.51p.p., 16.57p.p., and 7.18p.p., respectively, thus indicating the effectiveness of the algorithm.

Purpose Sensors are vital for autonomous driving, but they are susceptible to weather noise. To address the performance degradation of point cloud object detection in foggy conditions, this paper aims to propose an efficient dynamic pillar-based point cloud detection network tailored for foggy traffic scenarios. Design/methodology/approach The point cloud data are first pillarized, and then the model adaptively adjusts the feature extraction process through DynamicPillarVFE to significantly enhance the processing capability for sparse and irregular point cloud data. Second, a bird's eye view (BEV) backbone network based on residual connections is used to effectively mitigate the gradient vanishing problem in deep networks. In addition, this paper introduces Triplet Attention (TA) in the feature enhancement part, which can focus on important features more comprehensively and suppress noise interference. Findings In this paper, we mixed a small portion of the Multifog KITTI dataset on top of the KITTI dataset. Experimental results demonstrate that under moderate difficulty 3D evaluation metrics, our method achieves accuracy improvements of 5.03%, 6.66% and 7.78% for cars, pedestrians and cyclists, respectively, compared to the PointPillars baseline, significantly enhancing point cloud object detection performance in foggy conditions. Furthermore, the DPSF-Net architecture achieves an inference speed of 32.36 ms per frame, fully meeting the real-time processing requirements for autonomous driving applications. Originality/value This method is more economical to simulate the influence of sensors under foggy conditions, and through the training of mixed data sets, the network model can better cope with foggy interference.

In the recent 3D object detection methods for point clouds, the combination of point-based methods and voxel-based methods is gradually becoming a trend. Point-based methods retain the accurate position and pose information in the raw points and voxel-based methods get multi-scale structure information through the 3D backbone. However, because of the sparsity and irregularity of point clouds, both representations ignore the context information, which is important for the detection of sparse and small objects. To solve this problem, we propose a multi-stream feature aggregation network to extract features from three representations of the point cloud for object detection. Specifically, we exploit multi-stream features extracted from point, voxel, and perspective view (PV) respectively on a parallel way, where the complementary information between different perspectives can be used to enrich the feature representations, especially for the perspective view containing rich semantic context information. Secondly, to eliminate redundant information and better exploit the correlation between different feature representations, we design an attention-based multi-stream feature fusion module (MSFF) to combine features from three information streams. Besides, we introduce a new voxel RoI pooling with the self-attention in the second refinement stage, which can further strengthen the connection between local features in the proposal to obtain accurate classification and localization predictions. Our method achieves progressive results on the KITTI dataset, especially in the cyclist category, which improves the baseline significantly by 5.56%, 4.73%, 5.16% AP in the test set for easy, moderate, and hard levels respectively. Code will be available at https://github.com/june2678/MR F.

In this paper, we present a hardware-software implementation of a deep neural network for object detection based on a point cloud obtained by a LiDAR sensor. The Brevitas / PyTorch tools were used for network quantisation and the FINN tool for hardware implementation in the reprogrammable Zynq UltraScale+ MPSoC device. The PointPillars network was used in the research, as it is a reasonable compromise between detection accuracy and calculation complexity. The obtained results show that quite a significant computation precision limitation along with a few network architecture simplifications allows the solution to be implemented on an heterogeneous embedded platform with reasonable detection accuracy.

Accurate object detection is a fundamental requirement for autonomous systems to operate in dynamic urban environments. Considering the intricacy of the environment and the occlusion of objects, various point cloud based three-dimensional (3D) object detection methods have been proposed, such as point-based or voxel-based methods. In this paper, a feature selection mechanism is proposed in a voxel-based method to generate bird’s eye view (BEV) feature maps from the original point cloud. For the 3D Backbone, the U-shaped structure and single scale feature selection module are combined. After combining high-level semantics and low-level fine-grained features, the optimized features after the BEV feature maps are applied to region of interest (RoI) refinement, so that the voxel features can better serve the subsequent 3D object detection. The experimental results on KITTI dataset show higher 3D object detection accuracy compared to the state-of-the-art 3D detection methods, which reflects the effectiveness of the proposed architecture.

In the recent era, a lot of interest is attracted by the autonomous vehicles which can sense surroundings and navigate without human intervention. Object detection and recognition form a major part of autonomous driving systems. Lidar sensors can be used to capture point clouds of driving environment. Detecting multiple 3D objects in point clouds in real time and defining their boundaries with the help of 3D bounding boxes are critical in motion planning by self driving systems. This paper proposes a LiDAR-based 3D object detection system that operates in real-time, with emphasis on autonomous driving scenarios. A state-of-the-art 2D standard object detector for RGB images, YOLOv4, is used as the base for object detection. The multi-class 3D bounding boxes are generated using a complex regression approach. An Euler-Region-Proposal Network (E-RPN) is used to predict the pose of the object. The proposed model receives point cloud data as input and outputs 3D bounding boxes with classes in real-time. The experiments done on the KITTI benchmark dataset proves that the proposed system outperforms existing methods in terms of accuracy and performance.

3D object detection from raw and sparse point clouds has been far less treated to date, compared with its 2D counterpart. In this paper, we propose a novel framework called FVNet for 3D front-view proposal generation and object detection from point clouds. It consists of two stages: generation of front-view proposals and estimation of 3D bounding box parameters. Instead of generating proposals from camera images or bird's-eye-view maps, we first project point clouds onto a cylindrical surface to generate front-view feature maps which retains rich information. We then introduce a proposal generation network to predict 3D region proposals from the generated maps and further extrude objects of interest from the whole point cloud. Finally, we present another network to extract the point-wise features from the extruded object points and regress the final 3D bounding box parameters in the canonical coordinates. Our framework achieves real-time performance with 12ms per point cloud sample. Extensive experiments on the 3D detection benchmark KITTI show that the proposed architecture outperforms state-of-the-art techniques which take either camera images or point clouds as input, in terms of accuracy and inference time.

Currently, there are many kinds of voxel-based multisensor 3D object detectors, while point-based multisensor 3D object detectors have not been fully studied. In this paper, we propose a new 3D two-stage object detection method based on point cloud and image fusion to improve the detection accuracy. To address the problem of insufficient semantic information of point cloud, we perform multiscale deep fusion of LiDAR point and camera image in a point-wise manner to enhance point features. Due to the imbalance of LiDAR points, the object point cloud in the long-distance area is sparse. We design a point cloud completion module to predict the spatial shape of objects in the candidate boxes and extract the structural information to improve the feature representation ability to further refine the boxes. The framework is evaluated on widely used KITTI and SUN-RGBD dataset. Experimental results show that our method outperforms all state-of-the-art point-based 3D object detection methods and has comparable performance to voxel-based methods as well.

Segmenting or detecting objects in sparse Lidar point clouds are two important tasks in autonomous driving to allow a vehicle to act safely in its 3D environment. The best performing methods in 3D semantic segmentation or object detection rely on a large amount of annotated data. Yet annotating 3D Lidar data for these tasks is tedious and costly. In this context, we propose a self-supervised pretraining method for 3D perception models that is tailored to autonomous driving data. Specifically, we leverage the availability of synchronized and calibrated image and Lidar sensors in autonomous driving setups for distilling self-supervised pre-trained image representations into 3D models. Hence, our method does not require any point cloud nor image annotations. The keyingredient of our method is the use of superpixels which are used to pool 3D point features and 2D pixel features in visually similar regions. We then train a 3D network on the self-supervised task of matching these pooled point features with the corresponding pooled image pixel features. The advantages of contrasting regions obtained by superpixels are that: (1) grouping together pixels and points of visually coherent regions leads to a more meaningful contrastive task that produces features well adapted to 3D semantic segmentation and 3D object detection; (2) all the different regions have the same weight in the contrastive loss regardless of the number of 3D points sampled in these regions; (3) it mitigates the noise produced by incorrect matching of points and pixels due to occlusions between the different sensors. Extensive experiments on autonomous driving datasets demonstrate the ability of our image-to-Lidar distillation strategy to produce 3D representations that transfer well on semantic segmentation and object detection tasks.

Current point cloud object detection methods rely on expensive manual annotation. Utilizing contrastive learning for self-supervised pre-training on unlabeled large-scale point clouds can reduce annotation costs and improve model performance. However, selecting effective features for instance discrimination is crucial for contrastive learning. Previous methods have constructed instances for pre-training at different levels, such as points, proposals, and scenes, but the features of these instances differ from the objects to be detected. Considering that instance discrimination tasks based on object-level features align with downstream object detection tasks, we propose a novel and efficient self-supervised point cloud object detection pre-training framework called ObjectContrast. To learn more effective point cloud representations, this framework constructs two self-supervised pre-training modules: object-level instance discrimination contrast ObCo and bounding box geometric contrast prediction BoxCo . ObCo drives the model to learn general object representations to locate object foregrounds and determine categories. BoxCo enhances the model's geometric perception capabilities regarding the dimension and orientation of 3D bounding boxes. Extensive experiments on various detectors and datasets validate the efficiency and transferability of ObjectContrast. Compared with the state-of-the-art self-supervised pre-training methods, ObjectContrast demonstrates superior performance.

We propose octree-based transformers, named OctFormer, for 3D point cloud learning. OctFormer can not only serve as a general and effective backbone for 3D point cloud segmentation and object detection but also have linear complexity and is scalable for large-scale point clouds. The key challenge in applying transformers to point clouds is reducing the quadratic, thus overwhelming, computation complexity of attentions. To combat this issue, several works divide point clouds into non-overlapping windows and constrain attentions in each local window. However, the point number in each window varies greatly, impeding the efficient execution on GPU. Observing that attentions are robust to the shapes of local windows, we propose a novel octree attention, which leverages sorted shuffled keys of octrees to partition point clouds into local windows containing a fixed number of points while permitting shapes of windows to change freely. And we also introduce dilated octree attention to expand the receptive field further. Our octree attention can be implemented in 10 lines of code with open-sourced libraries and runs 17 times faster than other point cloud attentions when the point number exceeds 200 k. Built upon the octree attention, OctFormer can be easily scaled up and achieves state-of-the-art performances on a series of 3D semantic segmentation and 3D object detection benchmarks, surpassing previous sparse-voxel-based CNNs and point cloud transformers in terms of both efficiency and effectiveness. Notably, on the challenging ScanNet200 dataset, OctFormer outperforms sparse-voxel-based CNNs by 7.3 in mIoU. Our code and trained models are available at https://wang-ps.github.io/octformer.

Background: 3D object detection based on point clouds in road scenes has attracted much attention recently. The voxel-based methods voxelize the scene to regular grids, which can be processed with the advanced feature learning frameworks based on convolutional layers for semantic feature learning. The point-based methods can extract the geometric feature of the point due to the coordinate reservations. The combination of the two is effective for 3D object detection. However, the current methods use a voxel-based detection head with anchors for classification and localization. Although the preset anchors cover the entire scene, it is not suitable for detection tasks with larger scenes and multiple categories of objects, due to the limitation of the voxel size. Additionally, the misalignment between the predicted confidence and proposals in the Regions of the Interest (ROI) selection bring obstacles to 3D object detection. Methods: We investigate the combination of voxel-based methods and point-based methods for 3D object detection. Additionally, a voxel-to-point module that captures semantic and geometric features is proposed in the paper. The voxel-to-point module is conducive to the detection of small-size objects and avoids the presets of anchors in the inference stage. Moreover, a confidence adjustment module with the center-boundary-aware confidence attention is proposed to solve the misalignment between the predicted confidence and proposals in the regions of the interest selection. Results: The proposed method has achieved state-of-the-art results for 3D object detection in the Karlsruhe Institute of Technology and Toyota Technological Institute (KITTI) object detection dataset. Actually, as of September 19, 2021, our method ranked 1st in the 3D and Bird Eyes View (BEV) detection of cyclists tagged with difficulty level ‘easy’, and ranked 2nd in the 3D detection of cyclists tagged with ‘moderate’. Conclusions: We propose an end-to-end two-stage 3D object detector with voxel-to-point module and confidence adjustment module.