Graphical Models最新文献_第10页

Vertex position estimation with spatial–temporal transformer for 3D human reconstruction 基于时空变换的三维人体重构顶点位置估计

IF 1.7 4区计算机科学

Graphical Models Pub Date : 2023-12-01 Epub Date: 2023-10-26 DOI: 10.1016/j.gmod.2023.101207

Xiangjun Zhang, Yinglin Zheng, Wenjin Deng, Qifeng Dai, Yuxin Lin, Wangzheng Shi, Ming Zeng

{"title":"Vertex position estimation with spatial–temporal transformer for 3D human reconstruction","authors":"Xiangjun Zhang, Yinglin Zheng, Wenjin Deng, Qifeng Dai, Yuxin Lin, Wangzheng Shi, Ming Zeng","doi":"10.1016/j.gmod.2023.101207","DOIUrl":"https://doi.org/10.1016/j.gmod.2023.101207","url":null,"abstract":"<div><p>Reconstructing 3D human pose and body shape from monocular images or videos is a fundamental task for comprehending human dynamics. Frame-based methods can be broadly categorized into two fashions: those regressing parametric model parameters (e.g., SMPL) and those exploring alternative representations (e.g., volumetric shapes, 3D coordinates). Non-parametric representations have demonstrated superior performance due to their enhanced flexibility. However, when applied to video data, these non-parametric frame-based methods tend to generate inconsistent and unsmooth results. To this end, we present a novel approach that directly regresses the 3D coordinates of the mesh vertices and body joints with a spatial–temporal Transformer. In our method, we introduce a SpatioTemporal Learning Block (STLB) with Spatial Learning Module (SLM) and Temporal Learning Module (TLM), which leverages spatial and temporal information to model interactions at a finer granularity, specifically at the body token level. Our method outperforms previous state-of-the-art approaches on Human3.6M and 3DPW benchmark datasets.</p></div>","PeriodicalId":55083,"journal":{"name":"Graphical Models","volume":"130 ","pages":"Article 101207"},"PeriodicalIF":1.7,"publicationDate":"2023-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1524070323000371/pdfft?md5=a920877b3ee3210b23f7a6444d151f50&pid=1-s2.0-S1524070323000371-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"92047096","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Jrender: An efficient differentiable rendering library based on Jittor Jrender:一个基于Jittor的高效可微分渲染库

IF 1.7 4区计算机科学

Graphical Models Pub Date : 2023-12-01 Epub Date: 2023-10-18 DOI: 10.1016/j.gmod.2023.101202

Hanggao Xin, Chenzhong Xiang, Wenyang Zhou, Dun Liang

引用次数: 0

Modeling multi-style portrait relief from a single photograph 从一张照片建模多风格人像浮雕

IF 1.7 4区计算机科学

Graphical Models Pub Date : 2023-12-01 Epub Date: 2023-11-28 DOI: 10.1016/j.gmod.2023.101210

Yu-Wei Zhang , Hongguang Yang , Ping Luo , Zhi Li , Hui Liu , Zhongping Ji , Caiming Zhang

{"title":"Modeling multi-style portrait relief from a single photograph","authors":"Yu-Wei Zhang , Hongguang Yang , Ping Luo , Zhi Li , Hui Liu , Zhongping Ji , Caiming Zhang","doi":"10.1016/j.gmod.2023.101210","DOIUrl":"https://doi.org/10.1016/j.gmod.2023.101210","url":null,"abstract":"<div><p>This paper aims at extending the method of Zhang et al. (2023) to produce not only portrait bas-reliefs from single photographs, but also high-depth reliefs with reasonable depth ordering. We cast this task as a problem of style-aware photo-to-depth translation, where the input is a photograph conditioned by a style vector and the output is a portrait relief with desired depth style. To construct ground-truth data for network training, we first propose an optimization-based method to synthesize high-depth reliefs from 3D portraits. Then, we train a normal-to-depth network to learn the mapping from normal maps to relief depths. After that, we use the trained network to generate high-depth relief samples using the provided normal maps from Zhang et al. (2023). As each normal map has pixel-wise photograph, we are able to establish correspondences between photographs and high-depth reliefs. By taking the bas-reliefs of Zhang et al. (2023), the new high-depth reliefs and their mixtures as target ground-truths, we finally train a encoder-to-decoder network to achieve style-aware relief modeling. Specially, the network is based on a U-shaped architecture, consisting of Swin Transformer blocks to process hierarchical deep features. Extensive experiments have demonstrated the effectiveness of the proposed method. Comparisons with previous works have verified its flexibility and state-of-the-art performance.</p></div>","PeriodicalId":55083,"journal":{"name":"Graphical Models","volume":"130 ","pages":"Article 101210"},"PeriodicalIF":1.7,"publicationDate":"2023-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1524070323000401/pdfft?md5=de53c7cacd318b65effd57ea40c70f18&pid=1-s2.0-S1524070323000401-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138454034","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A systematic approach for enhancement of homogeneous background images using structural information 利用结构信息增强均匀背景图像的系统方法

IF 1.7 4区计算机科学

Graphical Models Pub Date : 2023-12-01 Epub Date: 2023-10-25 DOI: 10.1016/j.gmod.2023.101206

D. Vijayalakshmi , Malaya Kumar Nath

{"title":"A systematic approach for enhancement of homogeneous background images using structural information","authors":"D. Vijayalakshmi , Malaya Kumar Nath","doi":"10.1016/j.gmod.2023.101206","DOIUrl":"https://doi.org/10.1016/j.gmod.2023.101206","url":null,"abstract":"<div><p>Image enhancement is an indispensable pre-processing step for several image processing applications. Mainly, histogram equalization is one of the widespread techniques used by various researchers to improve the image quality by expanding the pixel values to fill the entire dynamic grayscale. It results in the visual artifact, structural information loss near edges due to the information loss (due to many-to-one mapping), and alteration in average luminance to a higher value. This paper proposes an enhancement algorithm based on structural information for homogeneous background images. The intensities are divided into two segments using the median value to preserve the average luminance. Unlike traditional techniques, this algorithm incorporates the spatial locations in the equalization process instead of the number of intensity values occurrences. The occurrences of each intensity concerning their spatial locations are combined using Rènyi entropy to enumerate a discrete function. An adaptive clipping limit is applied to the discrete function to control the enhancement rate. Then histogram equalization is performed on each segment separately, and the equalized segments are integrated to produce an enhanced image. The algorithm’s effectiveness is validated by evaluating the proposed method on CEED, CSIQ, LOL, and TID2013 databases. Experimental results reveal that the proposed method improves the contrast while preserving structural information, detail information, and average luminance. They are quantified by the high value of contrast improvement index, structural similarity index, and discrete entropy, and low value of average mean brightness error values of the proposed method when compared with the methods available in the literature, including deep learning architectures.</p></div>","PeriodicalId":55083,"journal":{"name":"Graphical Models","volume":"130 ","pages":"Article 101206"},"PeriodicalIF":1.7,"publicationDate":"2023-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S152407032300036X/pdfft?md5=66c749d2624c0d77acd46a4f2037626a&pid=1-s2.0-S152407032300036X-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"92047095","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

High-performance Ellipsoidal Clipmaps 高性能椭球体剪贴图

IF 1.7 4区计算机科学

Graphical Models Pub Date : 2023-12-01 Epub Date: 2023-11-30 DOI: 10.1016/j.gmod.2023.101209

Aleksandar Dimitrijević, Dejan Rančić

{"title":"High-performance Ellipsoidal Clipmaps","authors":"Aleksandar Dimitrijević, Dejan Rančić","doi":"10.1016/j.gmod.2023.101209","DOIUrl":"https://doi.org/10.1016/j.gmod.2023.101209","url":null,"abstract":"<div><p>This paper presents performance improvements for Ellipsoid Clipmaps, an out-of-core planet-sized geodetically accurate terrain rendering algorithm. The performance improvements were achieved by eliminating unnecessarily dense levels, more accurate block culling in the geographic coordinate system, and more efficient rendering methods. The elimination of unnecessarily dense levels is the result of analyzing and determining the optimal relative height of the viewer with respect to the most detailed level, resulting in the most consistent size of triangles across all visible levels. The proposed method for estimating the visibility of blocks based on view orientation allows rapid block-level view frustum culling performed in data space before visualization and spatial transformation of blocks. The use of a modern geometry pipeline through task and mesh shaders forced the handling of extremely fine granularity of blocks, but also shifted a significant part of the block culling process from CPU to the GPU. The approach described achieves high throughput and enables geodetically accurate rendering of the terrain based on the WGS 84 reference ellipsoid at very high resolution and in real time, with tens of millions of triangles with an average area of about 0.5 pix<span><math><msup><mrow></mrow><mrow><mn>2</mn></mrow></msup></math></span> on a 1080p screen on mid-range graphics cards.</p></div>","PeriodicalId":55083,"journal":{"name":"Graphical Models","volume":"130 ","pages":"Article 101209"},"PeriodicalIF":1.7,"publicationDate":"2023-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1524070323000395/pdfft?md5=26122c390b83d408f64d205c80bb4675&pid=1-s2.0-S1524070323000395-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138466486","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

PU-GAT: Point cloud upsampling with graph attention network PU-GAT:点云上采样与图关注网络

IF 1.7 4区计算机科学

Graphical Models Pub Date : 2023-12-01 Epub Date: 2023-09-25 DOI: 10.1016/j.gmod.2023.101201

Xuan Deng, Cheng Zhang, Jian Shi, Zizhao Wu

引用次数: 0

Fast progressive polygonal approximations for online strokes 快速渐进多边形近似在线笔画

IF 1.7 4区计算机科学

Graphical Models Pub Date : 2023-10-01 Epub Date: 2023-09-04 DOI: 10.1016/j.gmod.2023.101200

Mohammad Tanvir Parvez

引用次数: 1

RFMNet: Robust Deep Functional Maps for unsupervised non-rigid shape correspondence RFMNet:无监督非刚性形状对应的鲁棒深度函数映射

IF 1.7 4区计算机科学

Graphical Models Pub Date : 2023-10-01 Epub Date: 2023-07-29 DOI: 10.1016/j.gmod.2023.101189

Ling Hu , Qinsong Li , Shengjun Liu , Dong-Ming Yan , Haojun Xu , Xinru Liu

{"title":"RFMNet: Robust Deep Functional Maps for unsupervised non-rigid shape correspondence","authors":"Ling Hu , Qinsong Li , Shengjun Liu , Dong-Ming Yan , Haojun Xu , Xinru Liu","doi":"10.1016/j.gmod.2023.101189","DOIUrl":"https://doi.org/10.1016/j.gmod.2023.101189","url":null,"abstract":"<div><p>In traditional deep functional maps for non-rigid shape correspondence, estimating a functional map including high-frequency information requires enough linearly independent features via the least square method, which is prone to be violated in practice, especially at an early stage of training, or costly post-processing, e.g. ZoomOut. In this paper, we propose a novel method called RFMNet (<strong>R</strong>obust Deep <strong>F</strong>unctional <strong>M</strong>ap <strong>Net</strong>works), which jointly considers training stability and more geometric shape features than previous works. We directly first produce a pointwise map by resorting to optimal transport and then convert it to an initial functional map. Such a mechanism mitigates the requirements for the descriptor and avoids the training instabilities resulting from the least square solver. Benefitting from the novel strategy, we successfully integrate a state-of-the-art geometric regularization for further optimizing the functional map, which substantially filters the initial functional map. We show our novel computing functional map module brings more stable training even under encoding the functional map with high-frequency information and faster convergence speed. Considering the pointwise and functional maps, an unsupervised loss is presented for penalizing the correspondence distortion of Delta functions between shapes. To catch discretization-resistant and orientation-aware shape features with our network, we utilize DiffusionNet as a feature extractor. Experimental results demonstrate our apparent superiority in correspondence quality and generalization across various shape discretizations and different datasets compared to the state-of-the-art learning methods.</p></div>","PeriodicalId":55083,"journal":{"name":"Graphical Models","volume":"129 ","pages":"Article 101189"},"PeriodicalIF":1.7,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49889736","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

MixNet: Mix different networks for learning 3D implicit representations MixNet:混合不同的网络来学习3D隐式表示

IF 1.7 4区计算机科学

Graphical Models Pub Date : 2023-10-01 Epub Date: 2023-07-25 DOI: 10.1016/j.gmod.2023.101190

Bowen Lyu , Li-Yong Shen , Chun-Ming Yuan

引用次数: 0

Joint data and feature augmentation for self-supervised representation learning on point clouds 点云上自监督表示学习的联合数据和特征增强

IF 1.7 4区计算机科学

Graphical Models Pub Date : 2023-10-01 Epub Date: 2023-07-28 DOI: 10.1016/j.gmod.2023.101188

Zhuheng Lu , Yuewei Dai , Weiqing Li , Zhiyong Su

{"title":"Joint data and feature augmentation for self-supervised representation learning on point clouds","authors":"Zhuheng Lu , Yuewei Dai , Weiqing Li , Zhiyong Su","doi":"10.1016/j.gmod.2023.101188","DOIUrl":"https://doi.org/10.1016/j.gmod.2023.101188","url":null,"abstract":"<div><p>To deal with the exhausting annotations, self-supervised representation learning from unlabeled point clouds has drawn much attention, especially centered on augmentation-based contrastive methods. However, specific augmentations hardly produce sufficient transferability to high-level tasks on different datasets. Besides, augmentations on point clouds may also change underlying semantics. To address the issues, we propose a simple but efficient augmentation fusion contrastive learning framework to combine data augmentations in Euclidean space and feature augmentations in feature space. In particular, we propose a data augmentation method based on sampling and graph generation. Meanwhile, we design a data augmentation network to enable a correspondence of representations by maximizing consistency between augmented graph pairs. We further design a feature augmentation network that encourages the model to learn representations invariant to the perturbations using an encoder perturbation. We comprehensively conduct extensive object classification experiments and object part segmentation experiments to validate the transferability of the proposed framework. Experimental results demonstrate that the proposed framework is effective to learn the point cloud representation in a self-supervised manner, and yields state-of-the-art results in the community. The source code is publicly available at: <span>https://github.com/VCG-NJUST/AFSRL</span><svg><path></path></svg>.</p></div>","PeriodicalId":55083,"journal":{"name":"Graphical Models","volume":"129 ","pages":"Article 101188"},"PeriodicalIF":1.7,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49889741","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1