Cooperative perception for connected and automated vehicles is traditionally achieved through the fusion of feature maps from two or more vehicles. However, the absence of feature maps shared from other vehicles can lead to a significant decline in 3D object detection performance for cooperative perception models compared to standalone 3D detection models. This drawback impedes the adoption of cooperative perception as vehicle resources are often insufficient to concurrently employ two perception models. To tackle this issue, we present Simultaneous Individual and Cooperative Perception (SiCP), a generic framework that supports a wide range of the state-of-the-art standalone perception backbones and enhances them with a novel Dual-Perception Network (DP-Net) designed to facilitate both individual and cooperative perception. In addition to its lightweight nature with only 0.13M parameters, DP-Net is robust and retains crucial gradient information during feature map fusion. As demonstrated in a comprehensive evaluation on the V2V4Real and OPV2V datasets, thanks to DP-Net, SiCP surpasses state-of-the-art cooperative perception solutions while preserving the performance of standalone perception solutions. The source code can be found at https://github.com/DarrenQu/SiCP.

Vehicular cooperative perception aims to provide connected and automated vehicles (CAVs) with a longer and wider sensing range, making perception less susceptible to occlusions. However, this prospect is dimmed by the imperfection of onboard localization sensors such as Global Navigation Satellite Systems (GNSS), which can cause errors in aligning over-the-air perception data (from a remote vehicle) with a Host vehicle’s (HV’s) local observation. To mitigate this challenge, we propose a novel LiDAR-based relative localization framework based on the iterative closest point (ICP) algorithm. The framework seeks to estimate the correct transformation matrix between a pair of CAVs’ coordinate systems, through exchanging and matching a limited yet carefully chosen set of point clouds and usage of a coarse 2D map. From the deployment perspective, this means our framework only consumes conservative bandwidth in data transmission and can run efficiently with limited resources. Extensive evaluations on both synthetic dataset (COMAP) and KITTI-360 show that our proposed framework achieves state-of-the-art (SOTA) performance in cooperative localization. Therefore, it can be integrated with any upper-stream data fusion algorithm and serves as a preprocessor for high-quality cooperative perception.

Cooperative perception provides a novel way to conquer the sensing limitation on a single automated vehicle and potentially improves driving safety. To reduce the transmission data volume, existing solutions use the intermediate data generated by convolutional neural network (CNN) models, namely, feature maps, to achieve cooperative perception. The feature maps are however too large to be transmitted by the current V2X technology. We propose a novel approach, called Slim-FCP, to significantly reduce the transmission data size. It enables a channelwise feature encoder to remove irrelevant features for a better compression ratio. In addition, it adopts an intelligent channel selection strategy through which only representative channels of feature maps are selected for transmission. To evaluate the effectiveness of Slim-FCP, we further define a recall-to-bandwidth (RB) ratio metric to quantitatively measure how the recall of object detection changes with respect to the available network bandwidth. Experiment results show that Slim-FCP reduces the transmission data size by 75%, compared with the best state-of-the-art solution, with a subtle loss on object detection’s recall.

While self-attention mechanism has shown promising results for many vision tasks, it only considers the current features at a time. We show that such a manner cannot take full advantage of the attention mechanism. In this paper, we present Deep Connected Attention Network (DCANet), a novel design that boosts attention modules in a CNN model without any modification of the internal structure. To achieve this, we interconnect adjacent attention blocks, making information flow among attention blocks possible. With DCANet, all attention blocks in a CNN model are trained jointly, which improves the ability of attention learning. Our DCANet is generic. It is not limited to a specific attention module or base network architecture. Experimental results on ImageNet and MS COCO benchmarks show that DCANet consistently outperforms the state-of-the-art attention modules with a minimal additional computational overhead in all test cases. The code is available at: https://github.com/13952522076/DCANet.

To reduce the amount of transmitted data, feature map-based fusion is recently proposed as a practical solution to cooperative 3-D object detection by autonomous vehicles (AVs). The precision of object detection, however, may require significant improvement, especially for objects that are far away or occluded. To address this critical issue for the safety of AVs and human beings, we propose a cooperative spatial feature fusion (CoFF) method for AVs to effectively fuse feature maps for achieving a higher 3-D object detection performance. Especially, CoFF differentiates weights among feature maps for a more guided fusion, based on how much new semantic information is provided by the received feature maps. It also enhances the inconspicuous features corresponding to far/occluded objects to improve their detection precision. The experimental results show that CoFF achieves a significant improvement in terms of both detection precision and effective detection range for AVs, compared to previous feature fusion solutions.

Attention mechanisms have shown great success in computer vision. However, the commonly used global average pooling in some implementations aggregates a three-dimensional feature map to a one-dimensional attention map, leading a significant loss of structural information in the attention learning. In this article, we present a novel Spatial Pyramid Attention Network (SPANet), which exploits the structural information and channel relationships for better feature representation. SPANet enhances a base network by adding Spatial Pyramid Attention (SPA) blocks laterally. By rethinking the self-attention mechanism design, we further present three topology structures of attention path connection for our SPANet. They can be flexibly applied to various CNN architectures. SPANet is conceptually simple but practically powerful. It uses both structural regularization and structural information to achieve better learning capability. We have comprehensively evaluated the performance of SPANet on four benchmark datasets for different visual tasks. The experimental results show that SPANet significantly improves the recognition accuracy without adding much computation overhead. Using SPANet, we achieve an improvement of 1.6% top-1 classification accuracy on the ImageNet 2012 benchmark based on ResNet50, and SPANet outperforms SENet and other attention methods. SPANet also significantly improves the object detection performance by a clear margin with negligible additional computation overhead. When applying SPANet to RetinaNet based on the ResNet50 backbone, we improve the performance of the baseline model by 2.3 mAP and the enhanced model outperforms SENet and GCNet by 1.1 mAP and 1.7 mAP respectively. The code of SPANet is made publicly available.11[Online]. Available:https://github.com/13952522076/SPANet_TMM

In this paper, we present a cascaded context dependency module, which is a highly lightweight module that can improve the performance of deep convolutional neural networks for various visual tasks. Inspired by the feature pyramid work in object detection and the context dependency work in image recognition, we consider to cascade the contexts of multiscale feature maps to aggregate the locality and globality in a local region. We further extract the dependency between original input and cascaded contexts for feature recalibration. Without employing learnable layers, our method introduces almost no additional parameters and computations. Furthermore, Our module can be seamlessly plugged into many existing CNN architectures to improve the performance. Experiments on ImageNet and MS COCO benchmarks indicate that our method can achieve results on par with or better than related work. Qualitatively, we achieve an absolute 1.42% (77.3137% vs. 75.8974%) top-1 classification accuracy improvement based on ResNet50 on ImageNet 2012 validation set with negligible computational overhead. Besides, our method yields significant gains on the MS COCO benchmark for the object detection task. All codes and models are made publicly available1.1We submit all the codes, pre-trained models, and training log files to https://github.com/13952522076/ParameterFree.

Attention mechanism has shown great success in computer vision. In this paper, we introduce Spatial Pyramid Attention Network (SPANet) to investigate the role of attention block for image recognition. Our SPANet is conceptually simple but practically powerful. It enhances the base network by adding Spatial Pyramid Attention (SPA) Blocks laterally. In contrast to other attention based networks that leverage global average pooling, our proposed SPANet considers both structural regularization and structural information. Furthermore, we investigate the topology structure of attention path connection and present three SPANet structures. SPA block is flexible to be deployed to various convolutional neural network (CNN) architectures. The experimental results show that our SPANet significantly improves the recognition accuracy without introducing much computation overhead compared with other CNN models. Codes are made publicly available11https://github.com/13952522076/SPANet.

To make Convolutional Neural Networks (CNNs) more efficient and accurate, various lightweight self-attention modules have been proposed. In this paper, we systematically study state-of-the-art attention modules in CNNs and discover that self-attention mechanism can be closely related to normalization. Based on this observation, we propose a novel attention module, named Normalization-Attention module (NA module in short), which is almost parameter-free. The NA module calculates the mean and standard deviation of intermediate feature maps and processes the feature context with normalization, which makes a CNN model easier to be trained and more responsive to informative features. Our proposed Normalization-Attention module can be integrated into various base CNN architectures, and used for many computer vision tasks, including image recognition, object detection, and more. Experimental results on ImageNet and MS COCO benchmarks show that our method outperforms state-of-the-art works using fewer parameters. Codes are made publicly available.

Autonomous vehicles may make wrong decisions due to inaccurate detection and recognition. Therefore, an intelligent vehicle can combine its own data with that of other vehicles to enhance perceptive ability, and thus improve detection accuracy and driving safety. However, multi-vehicle cooperative perception requires the integration of real world scenes and the traffic of raw sensor data exchange far exceeds the bandwidth of existing vehicular networks. To the best our knowledge, we are the first to conduct a study on raw-data level cooperative perception for enhancing the detection ability of self-driving systems. In this work, relying on LiDAR 3D point clouds, we fuse the sensor data collected from different positions and angles of connected vehicles. A point cloud based 3D object detection method is proposed to work on a diversity of aligned point clouds. Experimental results on KITTI and our collected dataset show that the proposed system outperforms perception by extending sensing area, improving detection accuracy and promoting augmented results. Most importantly, we demonstrate it is possible to transmit point clouds data for cooperative perception via existing vehicular network technologies.

Autonomous driving vehicle relies heavily on its perception system to sense surrounding environments and make driving decisions. One important task on autonomous driving vehicles is to correctly recognize different traffic signs. However, the traffic signs in the wild can be in various conditions, e.g., occluded, deteriorated, or vandalized, and not all of them are recognizable. In this work, we propose a novel system that leverages the perception system on autonomous vehicle to identify occluded road signs in real time. Based on transfer learning, we propose the occluded sign classification network (OSCN) that is able to achieve a precision of 96.34% on a real-world dataset.

Water quality data is incredibly important and valuable, but its acquisition is not always trivial. A promising solution is to distribute a wireless sensor network in water to measure and collect the data; however, a drawback exists in that the batteries of the system must be replaced or recharged after being exhausted. To mitigate this issue, we designed a self-sustained water quality sensing system that is powered by renewable bioenergy generated from microbial fuel cells (MFCs). MFCs collect the energy released from native magnesium oxidizing microorganisms (MOMs) that are abundant in natural waters. The proposed energy-harvesting technology is environmentally friendly and can provide maintenance-free power to sensors for several years. Despite these benefits, an MFC can only provide microwatt-level power that is not sufficient to continuously power a sensor. To address this issue, we designed a power management module to accumulate energy when the input voltage is as low as 0.33V. We also proposed a radio-frequency (RF) activation technique to remotely activate sensors that otherwise are switched off in default. With this innovative technique, a sensor’s energy consumption in sleep mode can be completely avoided. Additionally, this design can enable on-demand data acquisitions from sensors. We implement the proposed system and evaluate its performance in a stream. In 3-month field experiments, we find the system is able to reliably collect water quality data and is robust to environment changes.

Massive trust assessment (MTA) in an Online Social Network (OSN), i.e., computing the trustworthiness of all users in the network, is crucial in various OSN-related applications. Existing solutions are either too slow or inaccurate in addressing the MTA problem. We propose the OpinionWalk algorithm that accurately and efficiently conducts MTA in an OSN. OpinionWalk models trust by the Dirichlet distribution and uses a matrix to represent the direct trust relations among users. From the perspective of a user, other users' trustworthiness are stored in a column vector that is iteratively updated when the algorithm “walks” through the network, in a breadth-first search manner. We identify the overlapping subproblems property in MTA and prove OpinionWalk is a more efficient solution. The accuracy and execution time of OpinionWalk are evaluated and compared to benchmark algorithms including EigenTrust, TrustRank, MoleTrust, TidalTrust and AssessTrust, using two real-world datasets (Advogato and Pretty Good Privacy). Experimental results indicate that OpinionWalk is an efficient and accurate solution to MTA, compared to previous algorithms.

Non-destructive concurrent transmissions recently attract widespread attention in wireless sensor networks research community, and many studies demonstrate that non-destructive interference in concurrent transmissions enables routing-free packet transmission with low latency and increased reliability. In this paper, we present a new MAC protocol, called Non-Destructive Interference MAC (NDI-MAC), that integrates non-destructive simultaneous transmissions into receiver-initiated protocols, achieving energy efficient and low latency data transmission under a variety of traffic loads. The capture effect is exploited in NDI-MAC to finish rendezvous between multiple senders and receivers. NDI-MAC also relies on triggercast, a distributed middleware to trigger synchronous packet transmissions with constructive interference. So as to ensure single-hop reliability, backcast primitive is used under unicast traffic. Evaluation results show that NDI-MAC achieves high performance in terms of energy consumption and data delivery latency under data dissemination and collection traffic.