Search

Your search keyword '"Chen, Qijun"' showing total 1,297 results
Start Over Author "Chen, Qijun"
1,297 results on '"Chen, Qijun"'

176. GeoPose: Dense Reconstruction Guided 6D Object Pose Estimation With Geometric Consistency

Author

Chen Qijun, Huiyi Chen, Zhou Guangliang, Wang Deming, and Yan Yi

Subjects
Pixel, business.industry, Computer science, Pipeline (computing), Normalization (image processing), RANSAC, Object (computer science), Computer Science Applications, Image (mathematics), Signal Processing, Media Technology, RGB color model, Computer vision, Artificial intelligence, Electrical and Electronic Engineering, business, Pose
Abstract

6D object pose estimation for texture-less objects from RGB images remains challenging, especially in occlusion scenarios. Instead of localizing sparse keypoints by regressing their image coordinates or heatmaps, which are sensitive to occlusion, we introduce GeoPose, a novel reconstruction guided pose estimation pipeline that predicts dense correspondences and leverages geometric consistency effectively. We first design a dense reconstruction network (ReconNet) to reconstruct pixel-wise object coordinates in normalization space. Dense 2D-3D correspondences are generated intuitively by our explicit pa-rameterization for 3D object models, which dismisses keypoint selection efforts. These 2D-3D correspondences are then utilized to estimate 6D poses by the PnP algorithm with RANSAC iterations. Furthermore, a novel Cycle Loss is proposed to provide 3D prior supervision, which significantly correlates with the pose estimation task by guiding geometric consistency between reconstruction (pixel to 3D) and reprojection (3D to pixel). In addition, a training data augmentation method is proposed to handle the insufficiency of 6D datasets, the acquisition of which is error-prone and time-consuming. Extensive experiments demonstrate that, compared with existing RGB-based methods, our GeoPose can achieve state-of-the-art (SOTA) 6D pose estimation performance on the LINEMOD, Occlusion LINEMOD and T-LESS datasets.

Published
2022

177. Improving Autonomous Vehicle Visual Perception by Fusing Human Gaze and Machine Vision

Author

Zhao, Yiyue, Lei, Cailin, Shen, Yu, Du, Yuchuan, and Chen, Qijun

Abstract

With high-definition sensors and sophisticated machine vision algorithms, the visual perception capability of autonomous vehicle (AV) has largely advanced. However, the visual perception performance of AVs may still be unstable in complex traffic environment. To improve the robustness and capability of risk detection of AV visual perception system, this work proposes a framework to fuse human gaze and the object detection results from vehicle vision based on the Laplacian Pyramid algorithm. We evaluate the proposed method on a level-2 AV to perceive the interactive vehicles at unsignalized intersections. Using Extended Kalman Filter, the trajectory of the human drivers’ gaze and the anchor boxes from AV object detection are fused. Results reveal that with human-vehicle visual fusion, the actual trajectory of interactive vehicles can be predicted more accurately than separately using human gaze or object detection algorithm. The findings show that human-vehicle visual fusion improves the perception accuracy and robustness of interactive objects in complex traffic environment. The method has the potential to enhance the attention mechanism of AV vision.

Published
2023
Full Text
View/download PDF

179. DL-PCN: Differential learning and parallel convolutional network for action recognition.

Author

Zeng, Qinyang, Dang, Ronghao, Fang, Qin, Liu, Chengju, and Chen, Qijun

Subjects
DEEP learning, HUMAN activity recognition, MACHINE learning, CONVOLUTIONAL neural networks, RECOGNITION (Psychology)
Abstract

Graph Convolution Network (GCN) algorithms have greatly improved the accuracy of skeleton-based human action recognition. GCN can utilize the spatial information between skeletal joints in subsequent frames better than other deep learning algorithms, which is beneficial for achieving high accuracy. However, the traditional GCN algorithms consume lots of computation for the stack of multiple primary GCN layers. Aiming at solving the problem, we introduce a lightweight network, a Differential Learning and Parallel Convolutional Networks (DL-PCN), whose key modules are Differential Learning (DLM) and the Parallel Convolutional Network (PCN). DLM features a feedforward connection, which carries the error information of GCN modules with the same structure, where GCN and CNN modules directly extract the original information from the input data, making the spatiotemporal information extracted by these modules more complete than that of GCN and CNN tandem structure. PCN comprises GCN and Convolution Neural Network (CNN) in parallel. Our network achieves comparable performance on the NTU RGB+D 60 dataset, the NTU RGB+D 120 dataset and the Northwestern-UCLA dataset while considering both accuracy and calculation parameters. [ABSTRACT FROM AUTHOR]

Published
2023
Full Text
View/download PDF

183. Urban Digital Twins for Intelligent Road Inspection

Author

Fan, Rui, primary, Zhang, Yikang, additional, Guo, Sicen, additional, Li, Jiahang, additional, Feng, Yi, additional, Su, Shuai, additional, Zhang, Yanting, additional, Wang, Wenshuo, additional, Jiang, Yu, additional, Junaid Bocus, Mohammud, additional, Zhu, Xingyi, additional, and Chen, Qijun, additional

Published
2022
Full Text
View/download PDF

184. Graphene quantum dots induce cascadic apoptosis via interaction with proteins associated with anti-oxidation after endocytosis by Trypanosoma brucei

Author

Xie, Yiwei, primary, Liang, Hongrui, additional, Jiang, Ning, additional, Liu, Dingyuan, additional, Zhang, Naiwen, additional, Li, Qilong, additional, Zhang, Kai, additional, Sang, Xiaoyu, additional, Feng, Ying, additional, Chen, Ran, additional, Zhang, Yiwei, additional, and Chen, Qijun, additional

Published
2022
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results