Zobrazeno 1 - 10
of 126
pro vyhledávání: '"Zhou Dingfu"'
Due to the domain differences and unbalanced disparity distribution across multiple datasets, current stereo matching approaches are commonly limited to a specific dataset and generalize poorly to others. Such domain shift issue is usually addressed
Externí odkaz:
http://arxiv.org/abs/2307.16509
Autor:
Fang, Jin, Zhou, Dingfu, Zhao, Jingjing, Wu, Chenming, Tang, Chulin, Xu, Cheng-Zhong, Zhang, Liangjun
Over the past few years, there has been remarkable progress in research on 3D point clouds and their use in autonomous driving scenarios has become widespread. However, deep learning methods heavily rely on annotated data and often face domain genera
Externí odkaz:
http://arxiv.org/abs/2301.12515
Autor:
Yin, Junbo, Fang, Jin, Zhou, Dingfu, Zhang, Liangjun, Xu, Cheng-Zhong, Shen, Jianbing, Wang, Wenguan
Dominated point cloud-based 3D object detectors in autonomous driving scenarios rely heavily on the huge amount of accurately labeled samples, however, 3D annotation in the point cloud is extremely tedious, expensive and time-consuming. To reduce the
Externí odkaz:
http://arxiv.org/abs/2207.12655
Autor:
Yin, Junbo, Zhou, Dingfu, Zhang, Liangjun, Fang, Jin, Xu, Cheng-Zhong, Shen, Jianbing, Wang, Wenguan
Existing approaches for unsupervised point cloud pre-training are constrained to either scene-level or point/voxel-level instance discrimination. Scene-level methods tend to lose local details that are crucial for recognizing the road objects, while
Externí odkaz:
http://arxiv.org/abs/2207.12654
While fine-tuning pre-trained networks has become a popular way to train image segmentation models, such backbone networks for image segmentation are frequently pre-trained using image classification source datasets, e.g., ImageNet. Though image clas
Externí odkaz:
http://arxiv.org/abs/2207.03335
3D point cloud registration in remote sensing field has been greatly advanced by deep learning based methods, where the rigid transformation is either directly regressed from the two point clouds (correspondences-free approaches) or computed from the
Externí odkaz:
http://arxiv.org/abs/2203.13239
Even though considerable progress has been made in deep learning-based 3D point cloud processing, how to obtain accurate correspondences for robust registration remains a major challenge because existing hard assignment methods cannot deal with outli
Externí odkaz:
http://arxiv.org/abs/2110.15250
Publikováno v:
In Optics and Laser Technology August 2024 175
Existing deep learning-based approaches for monocular 3D object detection in autonomous driving often model the object as a rotated 3D cuboid while the object's geometric shape has been ignored. In this work, we propose an approach for incorporating
Externí odkaz:
http://arxiv.org/abs/2108.11127
Accurate detection of obstacles in 3D is an essential task for autonomous driving and intelligent transportation. In this work, we propose a general multimodal fusion framework FusionPainting to fuse the 2D RGB image and 3D point clouds at a semantic
Externí odkaz:
http://arxiv.org/abs/2106.12449