Search

Your search keyword '"Chen, Qijun"' showing total 50 results
Start Over Author "Chen, Qijun" Publisher institute of electrical and electronics engineers (ieee)
50 results on '"Chen, Qijun"'

19. 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

21. A Novel Depth and Color Feature Fusion Framework for 6D Object Pose Estimation

Author

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

Subjects
business.industry, Computer science, Deep learning, Pipeline (computing), Feature extraction, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Point cloud, Pattern recognition, 02 engineering and technology, Computer Science Applications, Feature (computer vision), Signal Processing, 0202 electrical engineering, electronic engineering, information engineering, Media Technology, RGB color model, 020201 artificial intelligence & image processing, Point (geometry), Artificial intelligence, Electrical and Electronic Engineering, business, Pose
Abstract

This paper aims to solve the problem of estimating the 6D pose of an object under occlusion using RGB-D images. Most existing methods typically use the information of color and depth images separately to make predictions, which limits their performances in the presence of occlusion. Instead, we propose a pipeline to effectively fuse color and depth information and perform region-level pose estimation. Our method first uses a CNN to extract the color features, and then we obtain the fusion features by combining the color features into the point cloud. Unlike existing methods, the fusion features are in the form of point sets instead of feature maps. We further use a PointNet++-like network to process the fusion features, obtaining several region-level features. Each region-level feature can predict a pose with confidence. The pose with the highest confidence is chosen as the final output. Experiments show that the proposed method outperforms the state-of-the-art methods on both the LINEMOD and Occlusion LINEMOD datasets, indicating that the proposed pipeline can obtain accurate pose estimation results and is robust to occlusion.

Published
2021

22. Unsupervised Anomaly Segmentation Via Multilevel Image Reconstruction and Adaptive Attention-Level Transition

Author

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

Subjects
Computer science, business.industry, Feature extraction, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Pattern recognition, Iterative reconstruction, Image segmentation, Feature (computer vision), Robustness (computer science), Anomaly detection, Segmentation, Artificial intelligence, Electrical and Electronic Engineering, business, Instrumentation, Image restoration
Abstract

In product inspection scenarios, the objects have different characteristics of textures and different potential anomalies. Still, traditional methods cannot precisely segment structural anomalies and achieve optimal performance on all object categories. This article introduces a flexible multilevel image reconstruction (MLIR) framework for the unsupervised anomaly segmentation task. The MLIR integrates multiple generators with different downsampling ratios to serve as multiple reconstruction levels and leverage an adjustable perceptual similarity measurement to obtain the featurewise reconstruction errors, overcoming the challenge of detecting multiscale anomalies. Leveraging embeddings with different receptive fields for reconstruction, generators perform differently on the same target. Therefore, we depend on the reconstruction uncertainty to propose an adaptive attention-level transition (ALT) strategy that optimizes the detection performance automatically according to the target’s category. ALT adjusts the weights of both reconstruction levels and feature measurement scales simultaneously to utilize features at a consistent level for reconstruction and anomaly detection (AD). This strategy remarkably improves the system’s adaptivity for different applications. Moreover, we propose reducing the signal-to-noise ratio (SNR) of the images by adding pepper noise so that the reconstruction process is framed as an image denoising task, which improves feature extraction and image restoration robustness. We conduct extensive experiments on the MVTec AD dataset and demonstrate that the proposed method performs satisfactorily to different categories and achieves the state-of-the-art (SOTA) performance.

Published
2021

Catalog

Books, media, physical & digital resources
See catalog results