On-road Object Detection Research Articles

One Stage Monocular 3D Object Detection Utilizing Discrete Depth and Orientation Representation

On-road object detection is a critical component in an autonomous driving system. The safety of the vehicle can only be as good as the reliability of the on-road object detection system. Thus, developing a fast and robust object detection algorithm has been the primary goal of many automotive industries and institutes. In recent years, multi-purpose vision-based driver assistance systems have gained popularity with the emergence of a deep neural network. A monocular camera has been developed to locate an object in the image plane and estimate the distance of the said object in the real world or the vehicle plane. In this work, we present a monocular 3D object detection method that utilizes the discrete depth and orientation representation. Our proposed method strives to predict object locations on 3D space utilizing keypoint detection on the object’s center point. To improve the point detection, we employ center regression on the objects segmentation mask, reducing the detection offset significantly. The simplicity of our proposed network architecture and its one-stage approach allows our algorithm to achieve competitive speed compared with prior methods. Our proposed method is able to achieve 26.93% detection score on the Cityscapes 3D object detection dataset, outperforming the preceding monocular method by a margin of 2.8 points.

On-Road Object Detection and Tracking Based on Radar and Vision Fusion: A Review

Environment perception, one of the most fundamental and challenging problems of autonomous vehicles (AVs), has been widely studied in recent decades. Due to inferior fault tolerance and the insufficient information caused by a single autonomous sensor (e.g., radar, lidar, or camera), multisensor fusion plays a significant role in environment perception systems, and its performance directly defines the safety of AVs. Due to good performance and low cost, radar-vision (RV) fusion has become popular and widely applied in the mass production of AVs. However, there have been a few generalizations about RV fusion, and in that context, this article presents a comprehensive review on RV fusion for both object detection and object tracking by RV fusion. With respect to the input data and fusion framework, this article categorizes the existing fusion frameworks into two categories, providing a detailed overview of each: object detection and tracking by RV fusion. Also, the state-of-the-art detectors and trackers based on deep learning are introduced, along with an analysis of their advantages and limitations. Finally, challenges and improvements are summarized to facilitate future research in the RV fusion field.

Towards unified on-road object detection and depth estimation from a single image

Towards unified on-road object detection and depth estimation from a single image

Developing an On-Road Object Detection System Using Monovision and Radar Fusion

In this study, a millimeter-wave (MMW) radar and an onboard camera are used to develop a sensor fusion algorithm for a forward collision warning system. This study proposed integrating an MMW radar and camera to compensate for the deficiencies caused by relying on a single sensor and to improve frontal object detection rates. Density-based spatial clustering of applications with noise and particle filter algorithms are used in the radar-based object detection system to remove non-object noise and track the target object. Meanwhile, the two-stage vision recognition system can detect and recognize the objects in front of a vehicle. The detected objects include pedestrians, motorcycles, and cars. The spatial alignment uses a radial basis function neural network to learn the conversion relationship between the distance information of the MMW radar and the coordinate information in the image. Then a neural network is utilized for object matching. The sensor with a higher confidence index is selected as the system output. Finally, three kinds of scenario conditions (daytime, nighttime, and rainy-day) were designed to test the performance of the proposed method. The detection rates and the false alarm rates of proposed system were approximately 90.5% and 0.6%, respectively.

Road Object Detection Using a Disparity-Based Fusion Model

Detection methods based on 2-D images tend to extract the color, texture, shape, and other appearance features of objects. However, in complex scenes, the detection results using these methods are often influenced by shadows, occlusion, and resolution. In this paper, a disparity-proposal-based detection method that rapidly extracts candidate frames of the detection objects on the basis of stereo disparity and ensures the robustness of the candidate frames under different perturbations is proposed. Furthermore, depth information is used to construct multi-scale pooling layers, allowing objects of different sizes to activate different layers at different levels. The detection model incorporates 2-D image features and 3-D geometric features and overcomes the limitations of the 2-D detection methods (absence of depth information) by using disparity features. Based on the experimental results, this method effectively achieves on-road object detection in complex scenes.