no code implementations • 8 Apr 2024 • Fengrui Tian, Yaoyao Liu, Adam Kortylewski, Yueqi Duan, Shaoyi Du, Alan Yuille, Angtian Wang
Instead of using manually annotated images, we leverage diffusion models (e. g., Zero-1-to-3) to generate a set of images under controlled pose differences and propose to learn our object pose estimator with those images.
1 code implementation • 8 Apr 2024 • Fengrui Tian, Yueqi Duan, Angtian Wang, Jianfei Guo, Shaoyi Du
As there is 2D-to-3D ambiguity problem in the viewing direction when extracting 3D flow features from 2D video frames, we consider the volume densities as opacity priors that describe the contributions of flow features to the semantics on the frames.
no code implementations • 28 Mar 2024 • Yue Gao, Jiaxuan Lu, Siqi Li, Yipeng Li, Shaoyi Du
By treating segments as vertices and constructing hyperedges using rule-based and KNN-based strategies, a multi-view hypergraph neural network that captures relationships across viewpoint and temporal features is established.
1 code implementation • 27 Mar 2024 • Weidong Xie, Lun Luo, Nanfei Ye, Yi Ren, Shaoyi Du, Minhang Wang, Jintao Xu, Rui Ai, Weihao Gu, Xieyuanli Chen
Experimental results on the KITTI dataset show that our proposed methods achieve state-of-the-art performance while running in real time.
1 code implementation • 4 Jan 2024 • Xuehao Gao, Yang Yang, Zhenyu Xie, Shaoyi Du, Zhongqian Sun, Yang Wu
The whole text-driven human motion synthesis problem is then divided into multiple abstraction levels and solved with a multi-stage generation framework with a cascaded latent diffusion model: an initial generator first generates the coarsest human motion guess from a given text description; then, a series of successive generators gradually enrich the motion details based on the textual description and the previous synthesized results.
Ranked #6 on Motion Synthesis on KIT Motion-Language
no code implementations • 29 Jul 2023 • Yuehai Chen, Jing Yang, Badong Chen, Hua Gang, Shaoyi Du
To improve the robustness to annotation displacement, we design an effective transport cost function based on GGD.
1 code implementation • 5 May 2023 • Canhui Tang, Yiheng Li, Shaoyi Du, Guofa Wang, Zhiqiang Tian
Feature Descriptors and Detectors are two main components of feature-based point cloud registration.
1 code implementation • 29 Mar 2023 • Yiheng Li, Canhui Tang, Runzhao Yao, Aixue Ye, Feng Wen, Shaoyi Du
Firstly, we propose to use salient points with prominent local features as nodes to increase patch repeatability, and introduce some uniformly distributed points to complete the point cloud, thus constituting hybrid points.
no code implementations • CVPR 2023 • Xuehao Gao, Shaoyi Du, Yang Wu, Yang Yang
Encouraged by the effectiveness of encoding temporal dynamics within the frequency domain, recent human motion prediction systems prefer to first convert the motion representation from the original pose space into the frequency space.
1 code implementation • ICCV 2023 • Fengrui Tian, Shaoyi Du, Yueqi Duan
More specifically, we learn an implicit velocity field to estimate point trajectory from temporal features with Neural ODE, which is followed by a flow-based feature aggregation module to obtain spatial features along the point trajectory.
no code implementations • 21 Jun 2022 • Yuehai Chen, Jing Yang, Badong Chen, Shaoyi Du
Thus, CNN could locate and estimate crowds accurately in low-density regions, while it is hard to properly perceive the densities in high-density regions.
no code implementations • CVPR 2022 • Zhang Chen, Zhiqiang Tian, Jihua Zhu, Ce Li, Shaoyi Du
Our method is motivated by two cause-effect chains including category-causality chain and anatomy-causality chain.
no code implementations • 23 Jun 2021 • Yuehai Chen, Jing Yang, Dong Zhang, Kun Zhang, Badong Chen, Shaoyi Du
More specifically, we scan the whole input images and its priority maps in the form of column vector to obtain a relevance matrix estimating their similarity.
no code implementations • IEEE Access 2020 • YANG YANG, WEILE CHEN, Muyi Wang, DEXING ZHONG, Shaoyi Du
Different from traditional feature-based methods, we design a hybrid feature representation with color moments of the point, which could be applied naturally for any color point cloud.
no code implementations • 1 Jun 2017 • Congcong Jin, Jihua Zhu, Yaochen Li, Shaoyi Du, Zhongyu Li, Huimin Lu
For the registration of partially overlapping point clouds, this paper proposes an effective approach based on both the hard and soft assignments.