JZJun Zhang

Jun Zhang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Jun Zhang has a total of 9 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Lidar-camera range-view fusion for 3D object detection in autonomous driving

    by Xuzhong Hu, Zaipeng Duan, Pei An, Jun Zhang, Jie Ma


  • Open Access

    LiDAR-Camera Range-View fusion for 3D Object Detection in Autonomous Driving

    by Xuzhong Hu, Zaipeng Duan, Pei An, Jun Zhang, Jie Ma

    In recent years, LiDAR-Camera fusion for 3D object detection has emerged as a research hotspot, owing to its superior performance compared to single-sensor approaches. However, the vast majority of methods adopt fusion in the point, voxel, or BEV domains. The distinct representations of LiDAR-Camera data lead to domain discrepancies and misalignment of their features, resulting in suboptimal detection accuracy and slow inference speeds. In contrast, this paper introduces a novel range-view fusion strategy, leveraging its unique advantages --- data homogeneity, accurate feature alignment, and high efficiency --- over other methods. We propose the Range-view Pyramidal Fusion network (RPfusion), in which two modules are specifically designed to address the inherent problems of range-view fusion: the significant scale variations of objects and the feature vanishment of occluded objects. First, Appearance Feature Fusion (AFF) module adaptively integrates complementary geometric and semantic information across different scales using dilated cross-attention. Next, a Range-Guided Cross-Layer Modulation (RGCLM) module is proposed to extract rich contextual information for comprehensive perception and enhance features of occluded targets. The output features can globally decorate raw point cloud, serving as input to any LiDAR-based detectors. Additionally, we propose RPfusion-Det, which employs a RoIfusion module with grid-wise self-attention for local fusion, leading to more precise box refinement. Extensive experiments on the KITTI and NuScenes datasets demonstrate that RPfusion can consistently enhance various detection models. Meanwhile, RPfusion-Det achieves superior accuracy while maintaining efficient inference.

  • Two-point calibration method for a zoom camera with an approximate focal-invariant radial distortion model

    by Pei An, Jie Ma, Tao Ma, Bin Fang, Kun Yu, Xiaomao Liu, Jun Zhang

    A zoom camera can change its focal length and track moving objects with an adjustable resolution. To extract precise geometric information for the tracked objects, a zoom camera requires an accurate calibration method. High-precision camera calibration methods, however, usually require a number of control points that are not guaranteed in some practical situations. Most zoom cameras suffer radial distortion. Athough a traditional method can recover an undistorted image with known intrinsic parameters, it fails to work for a zoom camera with an unknown focal length. Motivated by these problems, we propose a two-point calibration method (TPCM). In this scheme, we first propose an approximate focal-invariant radial distortion (AFRD) model. With the AFRD model, an RGB image can be undistorted with an unknown focal length. After that, the TPCM method is presented to estimate the focal length and rotation matrix with only two control points for one image. Synthetic experiments demonstrate that the AFRD model is efficient. In the real data experiment, the mean reprojection error of the TPCM method is less than one pixel, which is smaller than current state-of-the-art methods, and we believe meets the demand for high-precision calibration.