Jianmin Ji

Jianmin Ji


Research Focus & Metrics

Academic Career Length

25Years (2002-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
17 May 2023 in International Journal of Computer Vision
Verified
Journal Article
Closed access
PDF file
YWYingjie Wang
Yingjie Wang
QMQiuyu Mao
Qiuyu Mao
HZHanqi Zhu
Hanqi Zhu
JDJiajun Deng
Jiajun Deng
YZYu Zhang
Yu Zhang
YZYanyong Zhang
Yanyong Zhang

11 July 2022 in IEEE Transactions on Multimedia
Verified
Journal Article
Closed access
HZHanqi Zhu
Hanqi Zhu
JDJiajun Deng
Jiajun Deng
YZYu Zhang
Yu Zhang
JJJianmin Ji
Jianmin Ji
QMQiuyu Mao
Qiuyu Mao
YZYanyong Zhang
Yanyong Zhang
It has been well recognized that fusing the complementary information from depth-aware LiDAR point clouds and semantic-rich stereo images would benefit 3D object detection. Nevertheless, it is non-trivial to explore the inherently unnatural interaction between sparse 3D points and dense 2D pixels. To ease this difficulty, the recent approaches generally project the 3D points onto the 2D image plane to sample the image data and then aggregate the data at the points. However, these approaches often suffer from the mismatch between the resolution of point clouds and RGB images, leading to sub-optimal performance. Specifically, taking the sparse points as the multi-modal data aggregation locations causes severe information loss for high-resolution images, which in turn undermines the effectiveness of multi-sensor fusion. In this paper, we present VPFNet—a new architecture that cleverly aligns and aggregates the point cloud and image data at the “virtual” points. Particularly, with their density lying between that of the 3D points and 2D pixels, the virtual points can nicely bridge the resolution gap between the two sensors, and thus preserve more information for processing. Moreover, we also investigate the data augmentation techniques that can be applied to both point clouds and RGB images, as the data augmentation has made non-negligible contribution towards 3D object detectors to date. We have conducted extensive experiments on KITTI dataset, and have observed good performance compared to the state-of-the-art methods. Remarkably, our VPFNet achieves 83.21% moderate $AP_{3D}$ and 91.86% moderate $AP_{BEV}$ on the KITTI test set. The network design also takes computation efficiency into consideration – we can achieve a FPS of 15 on a single NVIDIA RTX 2080Ti GPU. The source code is available at https://github.com/zhukevkesky/VPFNet.githttps://github.com/zhukevkesky/VPFNet.git.

23 November 2022 in Artificial Intelligence Review
Verified
Journal Article
Closed access
PDF file
XLXingchen Li
Xingchen Li
YXYuxuan Xiao
Yuxuan Xiao
BWBeibei Wang
Beibei Wang
HRHaojie Ren
Haojie Ren
YZYanyong Zhang
Yanyong Zhang
JJJianmin Ji
Jianmin Ji

25 July 2022 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
YLYao Li
Yao Li
JDJiajun Deng
Jiajun Deng
YZYu Zhang
Yu Zhang
JJJianmin Ji
Jianmin Ji
HLHouqiang Li
Houqiang Li
YZYanyong Zhang
Yanyong Zhang
A recent trend is to combine multiple sensors ( i.e. , cameras, LiDARs and millimeter-wave Radars) to achieve robust multi-modal perception for autonomous systems such as self-driving vehicles. Although quite a few sensor fusion algorithms have been proposed, some of which are top-ranked on various leaderboards, a systematic study on how to integrate these three types of sensors to develop effective multi-modal 3D object detection and tracking is still missing. Towards this end, we first study the strengths and weaknesses of each data modality carefully, and then compare several different fusion strategies to maximize their utility. Finally, based upon the lessons learnt, we propose a simple yet effective multi-modal 3D object detection and tracking framework (namely EZFusion). As demonstrated by extensive experiments on the nuScenes dataset, without fancy network modules, our proposed EZFusion makes remarkable improvements over the LiDAR-only baseline, and achieves comparable performance with the state-of-the-art fusion-based methods.

27 August 2020 in Sensors
Verified
Journal Article
Open access
PDF file
Guangda ChenGuangda Chen
Guangda Chen
SYShunyi Yao
Shunyi Yao
JMJun Ma
Jun Ma
LPLifan Pan
Lifan Pan
YCYu’an Chen
Yu’an Chen
XCXiao-Ping Chen
Xiao-Ping Chen
It is challenging to avoid obstacles safely and efficiently for multiple robots of different shapes in distributed and communication-free scenarios, where robots do not communicate with each other and only sense other robots’ positions and obstacles around them. Most existing multi-robot collision avoidance systems either require communication between robots or require expensive movement data of other robots, like velocities, accelerations and paths. In this paper, we propose a map-based deep reinforcement learning approach for multi-robot collision avoidance in a distributed and communication-free environment. We use the egocentric local grid map of a robot to represent the environmental information around it including its shape and observable appearances of other robots and obstacles, which can be easily generated by using multiple sensors or sensor fusion. Then we apply the distributed proximal policy optimization (DPPO) algorithm to train a convolutional neural network that directly maps three frames of egocentric local grid maps and the robot’s relative local goal positions into low-level robot control commands. Compared to other methods, the map-based approach is more robust to noisy sensor data, does not require robots’ movement data and considers sizes and shapes of related robots, which make it to be more efficient and easier to be deployed to real robots. We first train the neural network in a specified simulator of multiple mobile robots using DPPO, where a multi-stage curriculum learning strategy for multiple scenarios is used to improve the performance. Then we deploy the trained model to real robots to perform collision avoidance in their navigation without tedious parameter tuning. We evaluate the approach with multiple scenarios both in the simulator and on four differential-drive mobile robots in the real world. Both qualitative and quantitative experiments show that our approach is efficient and outperforms existing DRL-based approaches in many indicators. We also conduct ablation studies showing the positive effects of using egocentric grid maps and multi-stage curriculum learning.

11 April 2024 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
YLYao Li
Yao Li
YWYingjie Wang
Yingjie Wang
CMChengzhen Meng
Chengzhen Meng
YDYifan Duan
Yifan Duan
JJJianmin Ji
Jianmin Ji
YZYanyong Zhang
Yanyong Zhang
Far-range perception is essential for intelligent transportation systems. The main challenge of far-range perception is due to the difficulty of performing accurate object detection and tracking under far distances (e.g., $> \text{150}\,\text{m}$ ) at low cost. To cope with such challenges, deploying millimeter wave Radars and high-definition (HD) cameras, and fusing their data for joint perception has become a common practice. The key to this solution is the precise association between two types of data captured from different perspectives. Towards this goal, the first question is which plane to conduct the association, i.e., the 2D image plane or the BEV plane. We argue that the former is more suitable because the location errors of the perspective projection points are smaller at far distances and can lead to more accurate associations. Thus, we project Radar-based target locations from the BEV to the 2D plane and then associate them with camera-based object locations. Subsequently, we map the camera-based object locations to the BEV plane through inverse projection mapping (IPM) with corresponding depth information from Radar data. Finally, we engage a BEV tracking module to generate target trajectories for traffic monitoring. We devise a transformation parameters refining approach based on the depth scaling technique. We have deployed an actual testbed on an urban expressway and conducted extensive experiments for evaluation. The results show that our system can improve $\text{AP}_{\text{BEV}}$ by 32%, and reduce the location error by $\text{0.56}\,\text{m}$ . Our system is capable of achieving an average location accuracy of $\text{1.3}\,\text{m}$ within the $\text{500}\,\text{m}$ range.