no code implementations • 12 Nov 2024 • Liyuan Zhang, Le Hui, Qi Liu, Bo Li, Yuchao Dai
Multi-instance point cloud registration aims to estimate the pose of all instances of a model point cloud in the whole scene.
1 code implementation • 12 Sep 2024 • Yuan Wu, Zhiqiang Yan, Zhengxue Wang, Xiang Li, Le Hui, Jian Yang
MGHS projects the 2D image features into multiple subspaces, where each grid contains features within reasonable height ranges.
no code implementations • CVPR 2024 • Zhicheng Lu, Xiang Guo, Le Hui, Tianrui Chen, Min Yang, Xiao Tang, Feng Zhu, Yuchao Dai
In this way, our solution achieves 3D geometry-aware deformation modeling, which enables improved dynamic view synthesis and 3D dynamic reconstruction.
no code implementations • CVPR 2024 • YuFei Wang, Ge Zhang, Shaoqian Wang, Bo Li, Qi Liu, Le Hui, Yuchao Dai
In this paper we visualize the internal feature maps to analyze how the network densifies the input sparse depth.
1 code implementation • CVPR 2024 • Can Xu, Yuehui Han, Rui Xu, Le Hui, Jin Xie, Jian Yang
3D visual grounding aims to localize 3D objects described by free-form language sentences.
1 code implementation • 21 Dec 2023 • Yun Zhu, Le Hui, Yaqi Shen, Jin Xie
To this end, we propose a novel superpoint grouping network for indoor anchor-free one-stage 3D object detection.
Ranked #5 on 3D Object Detection on S3DIS
no code implementations • 1 Sep 2023 • Zhiqiang Yan, Xiang Li, Le Hui, Zhenyu Zhang, Jun Li, Jian Yang
To tackle these challenges, we explore a repetitive design in our image guided network to gradually and sufficiently recover depth values.
1 code implementation • CVPR 2023 • Yaqi Shen, Le Hui, Jin Xie, Jian Yang
In our superpoint generation module, we utilize the bidirectional flow information at the previous iteration to obtain the matching points of points and superpoint centers for soft point-to-superpoint association construction, in which the superpoints are generated for pairwise point clouds.
no code implementations • ICCV 2023 • Le Hui, Linghua Tang, Yuchao Dai, Jin Xie, Jian Yang
Then, to generate homogeneous superpoints from the sparse LiDAR point cloud, we propose a LiDAR point grouping algorithm that simultaneously considers the similarity of point embeddings and the Euclidean distance of points in 3D space.
1 code implementation • 11 Oct 2022 • Linghua Tang, Le Hui, Jin Xie
Due to the few annotated labels of 3D point clouds, how to learn discriminative features of point clouds to segment object instances is a challenging problem.
no code implementations • 14 Sep 2022 • Haobo Jiang, Kaihao Lan, Le Hui, Guangyu Li, Jin Xie, Jian Yang
The core of Siamese feature matching is how to assign high feature similarity on the corresponding points between the template and search area for precise object localization.
no code implementations • 9 Aug 2022 • Yikai Bian, Le Hui, Jianjun Qian, Jin Xie
Unsupervised domain adaptation for point cloud semantic segmentation has attracted great attention due to its effectiveness in learning with unlabeled data.
1 code implementation • 25 Jul 2022 • Mu He, Le Hui, Yikai Bian, Jian Ren, Jin Xie, Jian Yang
In this paper, we propose a resolution adaptive self-supervised monocular depth estimation method (RA-Depth) by learning the scale invariance of the scene depth.
1 code implementation • 25 Jul 2022 • Yuehui Han, Le Hui, Haobo Jiang, Jianjun Qian, Jin Xie
To this end, in this paper, we propose a novel adaptive subgraph generation based contrastive learning framework for efficient and robust self-supervised graph representation learning, and the optimal transport distance is utilized as the similarity metric between the subgraphs.
1 code implementation • 25 Jul 2022 • Le Hui, Lingpeng Wang, Linghua Tang, Kaihao Lan, Jin Xie, Jian Yang
Siamese network based trackers formulate 3D single object tracking as cross-correlation learning between point features of a template and a search area.
no code implementations • 24 Feb 2022 • Rui Xu, Zongyan Han, Le Hui, Jianjun Qian, Jin Xie
Then, we develop a generative adversarial network that combines the domain-specific features of the seen categories with the aligned domain-invariant features to synthesize samples, where the synthesized samples of the unseen categories are generated by using the corresponding word embeddings.
1 code implementation • 23 Feb 2022 • Yaqi Shen, Le Hui, Haobo Jiang, Jin Xie, Jian Yang
In this paper, we propose a neighborhood consensus based reliable inlier evaluation method for robust unsupervised point cloud registration.
1 code implementation • NeurIPS 2021 • Le Hui, Lingpeng Wang, Mingmei Cheng, Jin Xie, Jian Yang
The Siamese shape-aware feature learning network can capture 3D shape information of the object to learn the discriminative features of the object so that the potential target from the background in sparse point clouds can be identified.
1 code implementation • 1 Aug 2021 • Yifan Zhao, Le Hui, Jin Xie
To achieve this, we exploit the consistency between the input sparse point cloud and generated dense point cloud for the shapes and rendered images.
1 code implementation • 16 Apr 2021 • Mingmei Cheng, Le Hui, Jin Xie, Jian Yang
In order to reduce the number of annotated labels, we propose a semi-supervised semantic point cloud segmentation network, named SSPC-Net, where we train the semantic segmentation network by inferring the labels of unlabeled points from the few annotated 3D points.
1 code implementation • 7 Jan 2021 • Le Hui, Mingmei Cheng, Jin Xie, Jian Yang
In this paper, we develop an efficient point cloud learning network (EPC-Net) to form a global descriptor for visual place recognition, which can obtain good performance and reduce computation memory and inference time.
1 code implementation • ICCV 2021 • Le Hui, Hang Yang, Mingmei Cheng, Jin Xie, Jian Yang
In order to obtain discriminative global descriptors, we construct a pyramid VLAD module to aggregate the multi-scale feature maps of point clouds into the global descriptors.
Ranked #3 on 3D Place Recognition on Oxford RobotCar Dataset
1 code implementation • ICCV 2021 • Le Hui, Jia Yuan, Mingmei Cheng, Jin Xie, Xiaoya Zhang, Jian Yang
Specifically, in our clustering network, we first jointly learn a soft point-superpoint association map from the coordinate and feature spaces of point clouds, where each point is assigned to the superpoint with a learned weight.
no code implementations • 30 Jul 2020 • Mingmei Cheng, Le Hui, Jin Xie, Jian Yang, Hui Kong
In this paper, we propose a cascaded non-local neural network for point cloud segmentation.
1 code implementation • ECCV 2020 • Le Hui, Rui Xu, Jin Xie, Jianjun Qian, Jian Yang
Starting from the low-resolution point clouds, with the bilateral interpolation and max-pooling operations, the deconvolution network can progressively output high-resolution local and global feature maps.
5 code implementations • CVPR 2018 • Jifeng Wang, Xiang Li, Le Hui, Jian Yang
Specifically, a shadow image is fed into the first generator which produces a shadow detection mask.
Ranked #2 on Shadow Removal on SRD
no code implementations • 6 Dec 2017 • Le Hui, Xiang Li, Jiaxin Chen, Hongliang He, Chen Gong, Jian Yang
Unsupervised Image-to-Image Translation achieves spectacularly advanced developments nowadays.