Query:
学者姓名:孔德慧
Refining:
Year
Type
Indexed by
Source
Complex
Co-Author
Language
Clean All
Abstract :
Thanks to the development of 2D keypoint detectors, monocular 3D human pose estimation (HPE) via 2D-to-3D lifting approaches have achieved remarkable improvements. However, monocular 3D HPE is still a challenging problem due to the inherent depth ambiguities and occlusions. Recently, diffusion models have achieved great success in the field of image generation. Inspired by this, we transform 3D human pose estimation problem into a reverse diffusion process, and propose a dual-branch diffusion model so as to handle the indeterminacy and uncertainty of 3D pose and fully explore the global and local correlations between joints. Furthermore, we propose conditional dual-branch diffusion model to enhance the performance of 3D human pose estimation, in which the joint-level semantic information are regarded as the condition of the diffusion model, and integrated into the joint-level representations of 2D pose to enhance the expression of joints. The proposed method is verified on two widely used datasets and the experimental results have demonstrated the superiority.
Keyword :
Human pose estimation Human pose estimation Diffusion model Diffusion model Joint semantics Joint semantics Dual-branch Dual-branch
Cite:
Copy from the list or Export to your reference management。
GB/T 7714 | Li, Jinghua , Bai, Zhuowei , Kong, Dehui et al. 3d human pose estimation based on conditional dual-branch diffusion [J]. | MULTIMEDIA SYSTEMS , 2025 , 31 (1) . |
MLA | Li, Jinghua et al. "3d human pose estimation based on conditional dual-branch diffusion" . | MULTIMEDIA SYSTEMS 31 . 1 (2025) . |
APA | Li, Jinghua , Bai, Zhuowei , Kong, Dehui , Chen, Dongpan , Li, Qianxing , Yin, Baocai . 3d human pose estimation based on conditional dual-branch diffusion . | MULTIMEDIA SYSTEMS , 2025 , 31 (1) . |
Export to | NoteExpress RIS BibTex |
Abstract :
Human pose estimation based on monocular video has always been the focus of research in the human computer interaction community, which suffers mainly from depth ambiguity and self-occlusion challenges. While the recently proposed learning-based approaches have demonstrated promising performance, they do not fully explore the complementarity of features. In this paper, the authors propose a novel multi-feature and multi-level fusion network (MMF-Net), which extracts and combines joint features, bone features and trajectory features at multiple levels to estimate 3D human pose. In MMF-Net, firstly, the bone length estimation module and the trajectory multi-level fusion module are used to extract the geometric size information of the human body and multi-level trajectory information of human motion, respectively. Then, the fusion attention-based combination (FABC) module is used to extract multi-level topological structure information of the human body, and effectively fuse topological structure information, geometric size information and trajectory information. Extensive experiments show that MMF-Net achieves competitive results on Human3.6M, HumanEva-I and MPI-INF-3DHP datasets.
Keyword :
image processing image processing pose estimation pose estimation computer vision computer vision image reconstruction image reconstruction
Cite:
Copy from the list or Export to your reference management。
GB/T 7714 | Li, Qianxing , Kong, Dehui , Li, Jinghua et al. MMF-Net: A novel multi-feature and multi-level fusion network for 3D human pose estimation [J]. | IET COMPUTER VISION , 2025 , 19 (1) . |
MLA | Li, Qianxing et al. "MMF-Net: A novel multi-feature and multi-level fusion network for 3D human pose estimation" . | IET COMPUTER VISION 19 . 1 (2025) . |
APA | Li, Qianxing , Kong, Dehui , Li, Jinghua , Yin, Baocai . MMF-Net: A novel multi-feature and multi-level fusion network for 3D human pose estimation . | IET COMPUTER VISION , 2025 , 19 (1) . |
Export to | NoteExpress RIS BibTex |
Abstract :
弱监督时序动作定位任务的目标是在只有视频级标签的情况下,对未分割的视频中的动作进行分类和时序上的定位。目前基于神经网络模型的方法,大多训练分类器以预测视频片段级的类别分数,再融合其为视频级的类别分数。这些方法只关注视频的视觉特征,却忽视了视频语义结构信息。为进一步提升视频动作定位的质量,本文提出了一种视频语义结构信息辅助的弱监督时序动作定位方法。该方法首先以分类模块作为基础模型,然后基于视频在时序结构上的稀疏性和语义连续性等辅助信息设计一种平滑注意力模块,修正分类结果;另外,加入视频片段级语义标签预测模块,改善弱监督标签信息不充足问题;最后将三个模块共同训练以融合提升时序动作定位的精度。通过在THUMOS14和ActivityNet数据集上的实验,表明本文方法的性能指标明显优于目前现有方法。
Keyword :
语义结构信息 语义结构信息 伪标签 伪标签 注意力值 注意力值 动作定位 动作定位 弱监督 弱监督
Cite:
Copy from the list or Export to your reference management。
GB/T 7714 | 孔德慧 , 许梦文 , 李敬华 et al. 一种视频语义结构信息辅助的弱监督时序动作定位方法 [C] //2021中国自动化大会论文集 . 2022 . |
MLA | 孔德慧 et al. "一种视频语义结构信息辅助的弱监督时序动作定位方法" 2021中国自动化大会论文集 . (2022) . |
APA | 孔德慧 , 许梦文 , 李敬华 , 王少帆 , 尹宝才 . 一种视频语义结构信息辅助的弱监督时序动作定位方法 2021中国自动化大会论文集 . (2022) . |
Export to | NoteExpress RIS BibTex |
Abstract :
3D pose estimation remains a challenging task since human poses exhibit high ambiguity and multigranularity. Traditional graph convolution networks (GCNs) accomplish the task by modeling all skeletons as an entire graph, and are unable to fuse combinable part-based features. By observing that human movements occur due to part of human body (i.e. related skeletons and body components, known as the poselet) and those poselets contribute to each movement in a hierarchical fashion, we propose a hierarchical poselet-guided graph convolutional network (HPGCN) for 3D pose estimation from 2D poses. HPGCN sets five primitives of human body as basic poselets, and constitutes high-level poselets according to the kinematic configuration of human body. Moreover, HPGCN forms a fundamental unit by using a diagonally dominant graph convolution layer and a non-local layer, which corporately capture the multi-granular feature of human poses from local to global perspective. Finally HPGCN designs a geometric constraint loss function with constraints on lengths and directions of bone vectors, which help produce reasonable pose regression. We verify the effectiveness of HPGCN on three public 3D human pose benchmarks. Experimental results show that HPGCN outperforms several state-of-the-art methods. (c) 2021 Elsevier B.V.
Keyword :
Graph convolutional network Graph convolutional network Hierarchical poselet Hierarchical poselet Geometric constraint Geometric constraint Diagonally dominant graph convolution Diagonally dominant graph convolution Pose estimation Pose estimation
Cite:
Copy from the list or Export to your reference management。
GB/T 7714 | Wu, Yongpeng , Kong, Dehui , Wang, Shaofan et al. HPGCN: Hierarchical poselet-guided graph convolutional network for 3D pose estimation [J]. | NEUROCOMPUTING , 2022 , 487 : 243-256 . |
MLA | Wu, Yongpeng et al. "HPGCN: Hierarchical poselet-guided graph convolutional network for 3D pose estimation" . | NEUROCOMPUTING 487 (2022) : 243-256 . |
APA | Wu, Yongpeng , Kong, Dehui , Wang, Shaofan , Li, Jinghua , Yin, Baocai . HPGCN: Hierarchical poselet-guided graph convolutional network for 3D pose estimation . | NEUROCOMPUTING , 2022 , 487 , 243-256 . |
Export to | NoteExpress RIS BibTex |
Abstract :
Featured Application The vehicle detection algorithm proposed in this work could be used in autonomous driving systems to understand the environment, or could be applied in surveillance systems to extract useful transportation information through a camera. Vehicle detection as a special case of object detection has practical meaning but faces challenges, such as the difficulty of detecting vehicles of various orientations, the serious influence from occlusion, the clutter of background, etc. In addition, existing effective approaches, like deep-learning-based ones, demand a large amount of training time and data, which causes trouble for their application. In this work, we propose a dictionary-learning-based vehicle detection approach which explicitly addresses these problems. Specifically, an ensemble of sparse-and-dense dictionaries (ESDD) are learned through supervised low-rank decomposition; each pair of sparse-and-dense dictionaries (SDD) in the ensemble is trained to represent either a subcategory of vehicle (corresponding to certain orientation range or occlusion level) or a subcategory of background (corresponding to a cluster of background patterns) and only gives good reconstructions to samples of the corresponding subcategory, making the ESDD capable of classifying vehicles from background even though they exhibit various appearances. We further organize ESDD into a two-level cascade (CESDD) to perform coarse-to-fine two-stage classification for better performance and computation reduction. The CESDD is then coupled with a downstream AdaBoost process to generate robust classifications. The proposed CESDD model is used as a window classifier in a sliding-window scan process over image pyramids to produce multi-scale detections, and an adapted mean-shift-like non-maximum suppression process is adopted to remove duplicate detections. Our CESDD vehicle detection approach is evaluated on KITTI dataset and compared with other strong counterparts; the experimental results exhibit the effectiveness of CESDD-based classification and detection, and the training of CESDD only demands small amount of time and data.
Keyword :
ensemble learning ensemble learning object detection object detection dictionary learning dictionary learning vehicle detection vehicle detection
Cite:
Copy from the list or Export to your reference management。
GB/T 7714 | Rong, Zihao , Wang, Shaofan , Kong, Dehui et al. A Cascaded Ensemble of Sparse-and-Dense Dictionaries for Vehicle Detection [J]. | APPLIED SCIENCES-BASEL , 2021 , 11 (4) . |
MLA | Rong, Zihao et al. "A Cascaded Ensemble of Sparse-and-Dense Dictionaries for Vehicle Detection" . | APPLIED SCIENCES-BASEL 11 . 4 (2021) . |
APA | Rong, Zihao , Wang, Shaofan , Kong, Dehui , Yin, Baocai . A Cascaded Ensemble of Sparse-and-Dense Dictionaries for Vehicle Detection . | APPLIED SCIENCES-BASEL , 2021 , 11 (4) . |
Export to | NoteExpress RIS BibTex |
Abstract :
为了实现更为精确的视频目标跟踪,提出一种以时空上下文模型为基础的RGB-D序列目标跟踪算法.通过引入更新模板的深度信息,该模型精准地区分了输入序列的目标区域与背景区域,实现了深度权值和颜色权值的融合;基于目标序列的深度及目标动量计算,该模型有效地实现了尺度更新与遮挡处理.通过在RGB-D图像序列数据集上的详细实验评估,该时空上下文模型相对于其他先进的同类方法表现出更好的性能.因此,该方法实现了更为精确可靠的视频目标跟踪.
Keyword :
目标跟踪 目标跟踪 机器学习 机器学习 RGB-D RGB-D 目标动量 目标动量 计算机视觉 计算机视觉 时空上下文 时空上下文
Cite:
Copy from the list or Export to your reference management。
GB/T 7714 | 孔德慧 , 荣子豪 , 贾思宇 et al. 基于时空上下文模型的RGB-D序列目标跟踪方法 [J]. | 北京工业大学学报 , 2021 , 47 (03) : 224-230 . |
MLA | 孔德慧 et al. "基于时空上下文模型的RGB-D序列目标跟踪方法" . | 北京工业大学学报 47 . 03 (2021) : 224-230 . |
APA | 孔德慧 , 荣子豪 , 贾思宇 , 王少帆 , 尹宝才 . 基于时空上下文模型的RGB-D序列目标跟踪方法 . | 北京工业大学学报 , 2021 , 47 (03) , 224-230 . |
Export to | NoteExpress RIS BibTex |
Abstract :
基于时空上下文模型的RGB-D序列目标跟踪方法
Keyword :
目标跟踪 目标跟踪 时空上下文 时空上下文 计算机视觉 计算机视觉 RGB-D RGB-D 目标动量 目标动量 机器学习 机器学习
Cite:
Copy from the list or Export to your reference management。
GB/T 7714 | 孔德慧 , 荣子豪 , 贾思宇 et al. 基于时空上下文模型的RGB-D序列目标跟踪方法 [J]. | 孔德慧 , 2021 , 47 (3) : 224-230 . |
MLA | 孔德慧 et al. "基于时空上下文模型的RGB-D序列目标跟踪方法" . | 孔德慧 47 . 3 (2021) : 224-230 . |
APA | 孔德慧 , 荣子豪 , 贾思宇 , 王少帆 , 尹宝才 , 北京工业大学学报 . 基于时空上下文模型的RGB-D序列目标跟踪方法 . | 孔德慧 , 2021 , 47 (3) , 224-230 . |
Export to | NoteExpress RIS BibTex |
Abstract :
Sparse representation is a powerful tool in many visual applications since images can be represented effectively and efficiently with a dictionary. Conventional dictionary learning methods usually treat each training sample equally, which would lead to the degradation of recognition performance when the samples from same category distribute dispersedly. This is because the dictionary focuses more on easy samples (known as highly clustered samples), and those hard samples (known as widely distributed samples) are easily ignored. As a result, the test samples which exhibit high dissimilarities to most of intra-category samples tend to be misclassified. To circumvent this issue, this paper proposes a simple and effective hardness-aware dictionary learning (HADL) method, which considers training samples discriminatively based on the AdaBoost mechanism. Different from learning one optimal dictionary, HADL learns a set of dictionaries and corresponding sub-classifiers jointly in an iterative fashion. In each iteration, HADL learns a dictionary and a sub-classifier, and updates the weights based on the classification errors given by current sub-classifier. Those correctly classified samples are assigned with small weights while those incorrectly classified samples are assigned with large weights. Through the iterated learning procedure, the hard samples are associated with different dictionaries. Finally, HADL combines the learned sub-classifiers linearly to form a strong classifier, which improves the overall recognition accuracy effectively. Experiments on well-known benchmarks show that HADL achieves promising classification results.
Keyword :
Dictionaries Dictionaries Training Training classification classification Boosting Boosting Visualization Visualization AdaBoost AdaBoost dictionary learning dictionary learning Task analysis Task analysis Face recognition Face recognition Sparse representation Sparse representation
Cite:
Copy from the list or Export to your reference management。
GB/T 7714 | Wang, Lichun , Li, Shuang , Wang, Shaofan et al. Hardness-Aware Dictionary Learning: Boosting Dictionary for Recognition [J]. | IEEE TRANSACTIONS ON MULTIMEDIA , 2021 , 23 : 2857-2867 . |
MLA | Wang, Lichun et al. "Hardness-Aware Dictionary Learning: Boosting Dictionary for Recognition" . | IEEE TRANSACTIONS ON MULTIMEDIA 23 (2021) : 2857-2867 . |
APA | Wang, Lichun , Li, Shuang , Wang, Shaofan , Kong, Dehui , Yin, Baocai . Hardness-Aware Dictionary Learning: Boosting Dictionary for Recognition . | IEEE TRANSACTIONS ON MULTIMEDIA , 2021 , 23 , 2857-2867 . |
Export to | NoteExpress RIS BibTex |
Abstract :
Three-dimensional (3D) reconstruction of shapes is an important research topic in the fields of computer vision, computer graphics, pattern recognition, and virtual reality. Existing 3D reconstruction methods usually suffer from two bottlenecks: (1) they involve multiple manually designed states which can lead to cumulative errors, but can hardly learn semantic features of 3D shapes automatically; (2) they depend heavily on the content and quality of images, as well as precisely calibrated cameras. As a result, it is difficult to improve the reconstruction accuracy of those methods. 3D reconstruction methods based on deep learning overcome both of these bottlenecks by automatically learning semantic features of 3D shapes from low-quality images using deep networks. However, while these methods have various architectures, in-depth analysis and comparisons of them are unavailable so far. We present a comprehensive survey of 3D reconstruction methods based on deep learning. First, based on different deep learning model architectures, we divide 3D reconstruction methods based on deep learning into four types, recurrent neural network, deep autoencoder, generative adversarial network, and convolutional neural network based methods, and analyze the corresponding methodologies carefully. Second, we investigate four representative databases that are commonly used by the above methods in detail. Third, we give a comprehensive comparison of 3D reconstruction methods based on deep learning, which consists of the results of different methods with respect to the same database, the results of each method with respect to different databases, and the robustness of each method with respect to the number of views. Finally, we discuss future development of 3D reconstruction methods based on deep learning.
Keyword :
Deep learning models Deep learning models Deep autoencoder Deep autoencoder Convolutional neural network Convolutional neural network Recurrent neural network Recurrent neural network TP391 TP391 Generative adversarial network Generative adversarial network Three-dimensional reconstruction Three-dimensional reconstruction
Cite:
Copy from the list or Export to your reference management。
GB/T 7714 | Liu, Caixia , Kong, Dehui , Wang, Shaofan et al. Deep3D reconstruction: methods, data, and challenges [J]. | FRONTIERS OF INFORMATION TECHNOLOGY & ELECTRONIC ENGINEERING , 2021 , 22 (5) : 652-672 . |
MLA | Liu, Caixia et al. "Deep3D reconstruction: methods, data, and challenges" . | FRONTIERS OF INFORMATION TECHNOLOGY & ELECTRONIC ENGINEERING 22 . 5 (2021) : 652-672 . |
APA | Liu, Caixia , Kong, Dehui , Wang, Shaofan , Wang, Zhiyong , Li, Jinghua , Yin, Baocai . Deep3D reconstruction: methods, data, and challenges . | FRONTIERS OF INFORMATION TECHNOLOGY & ELECTRONIC ENGINEERING , 2021 , 22 (5) , 652-672 . |
Export to | NoteExpress RIS BibTex |
Abstract :
三维形状重建是计算机视觉、计算机图形学、模式识别和虚拟现实等领域的重要研究课题。现有三维重建方法通常存在两个瓶颈:(1)它们涉及多个人工设计阶段,导致累积误差,且难以自动学习三维形状的语义特征;(2)它们严重依赖图像内容和质量,以及精确校准的摄像机。因此,这些方法的重建精度难以提高。基于深度学习的三维重建方法通过利用深度网络自动学习低质量图像中的三维形状语义特征,克服了这两个瓶颈。然而,这些方法具有多种体系框架,但是至今未有文献对它们作深入分析和比较。本文对基于深度学习的三维重建方法进行全面综述。首先,基于不同深度学习模型框架,将基于深度学习的三维重建方法分为4类:递归神经网络、深自编码器、生...
Keyword :
生成对抗网络 生成对抗网络 循环神经网络 循环神经网络 三维重建 三维重建 深度学习模型 深度学习模型 深度自编码器 深度自编码器 卷积神经网络 卷积神经网络
Cite:
Copy from the list or Export to your reference management。
GB/T 7714 | 刘彩霞 , 孔德慧 , 王少帆 et al. 深度三维重建:方法、数据和挑战(英文) [J]. | Frontiers of Information Technology & Electronic Engineering , 2021 , 22 (05) : 652-673 . |
MLA | 刘彩霞 et al. "深度三维重建:方法、数据和挑战(英文)" . | Frontiers of Information Technology & Electronic Engineering 22 . 05 (2021) : 652-673 . |
APA | 刘彩霞 , 孔德慧 , 王少帆 , 王志勇 , 李敬华 , 尹宝才 . 深度三维重建:方法、数据和挑战(英文) . | Frontiers of Information Technology & Electronic Engineering , 2021 , 22 (05) , 652-673 . |
Export to | NoteExpress RIS BibTex |
Export
Results: |
Selected to |
Format: |