Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry最新文献

WLA-Net: A Whole New Light-weight Architecture For Visual Task WLA-Net:一个全新的用于视觉任务的轻量级架构

Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry Pub Date : 2022-12-27 DOI: 10.1145/3574131.3574458

Liuhao Yu, Danyang Yao, Mei Fang, Lei Zhu

{"title":"WLA-Net: A Whole New Light-weight Architecture For Visual Task","authors":"Liuhao Yu, Danyang Yao, Mei Fang, Lei Zhu","doi":"10.1145/3574131.3574458","DOIUrl":"https://doi.org/10.1145/3574131.3574458","url":null,"abstract":"In this paper, we introduce WLA-Net, a whole new convolutional networks that have smaller parameters and FLOPs model. WLA-Net are based on a cross architecture that uses mechanism of attention and Residual block to build light deep neural networks. While improving the classification accuracy, the parameters of model is reduced, make the model more lightweight and improving resource utilization. A lightweight convolution module is designed in the network that can perform image classification tasks accurately and efficiently while introducing a module that large Convolution attention to improve image classification accuracy. In addition, an new AttentionModule is proposed, which mines information aggregations in the channel direction as much as possible to extract more efficient depth features. It can effectively fuse the features of the channels in the image to obtain higher accuracy. At the same time, a new residual structure is designed to fuse the information between feature channels to make it more closely related. The image classification accuracy of the model is verified on the large natural images datasets. Experimental results show that the proposed method has SOTA performance.","PeriodicalId":111802,"journal":{"name":"Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry","volume":"102 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122939947","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Spatial-temporal Fusion Network for Fast Video Shadow Detection 快速视频阴影检测的时空融合网络

Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry Pub Date : 2022-12-27 DOI: 10.1145/3574131.3574455

Jun-Hong Lin, Liansheng Wang

引用次数: 0

Envisioning A Hyper-Learning System in the Age of Metaverse 设想一个超宇宙时代的超级学习系统

Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry Pub Date : 2022-12-27 DOI: 10.1145/3574131.3574427

Anqi Wang, Ze-Feng Gao, Zeyu Wang, Pan Hui, Tristan Braud

引用次数: 1

Exploring the Effect of Navigation Tool Design on Virtual Environment’s Navigation and Revisiting Experience 探索导航工具设计对虚拟环境导航和重游体验的影响

Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry Pub Date : 2022-12-27 DOI: 10.1145/3574131.3574447

Chengsi Yi, Yi-Chieh Lee

{"title":"Exploring the Effect of Navigation Tool Design on Virtual Environment’s Navigation and Revisiting Experience","authors":"Chengsi Yi, Yi-Chieh Lee","doi":"10.1145/3574131.3574447","DOIUrl":"https://doi.org/10.1145/3574131.3574447","url":null,"abstract":"An effective navigation tool is critical for traversing a spatially ambiguous virtual environment. The navigation tool should also prompt the gain of spatial knowledge of the current environment, which supports the later revisit. Regarding the landmark-route-survey theory and path integration theory, we designed a dynamic-lines navigation tool incorporating multiple navigation lines. We examined it in a multi-floor ambiguous environment against a map-based navigation tool. Our results show that the dynamic-lines navigation tool outperformed the map navigation tool with less completion time and better user experience reported while performing the same as the map in later unassisted revisits. Besides, we found that subjects were more prone to encode goal positions with sequential landmarks even in a landmark-less environment. This finding implies that such preference might result from minimal bodily sensories when interacting with a computer-based experiment platform, suggesting that future virtual-navigation design should consider either compensating for the lack of bodily information or supporting the sequential encoding preference of spatial information.","PeriodicalId":111802,"journal":{"name":"Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125993228","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

MMTrans: MultiModal Transformer for realistic video virtual try-on MMTrans:多模态变压器的现实视频虚拟试戴

Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry Pub Date : 2022-12-27 DOI: 10.1145/3574131.3574431

Xinrong Hu, Ziyi Zhang, Ruiqi Luo, Junjie Huang, Jinxing Liang, Jin Huang, Tao Peng, Hao Cai

{"title":"MMTrans: MultiModal Transformer for realistic video virtual try-on","authors":"Xinrong Hu, Ziyi Zhang, Ruiqi Luo, Junjie Huang, Jinxing Liang, Jin Huang, Tao Peng, Hao Cai","doi":"10.1145/3574131.3574431","DOIUrl":"https://doi.org/10.1145/3574131.3574431","url":null,"abstract":"Video virtual try-on methods aim to generate coherent, smooth, and realistic try-on videos, it matches the target clothing with the person in the video in a spatiotemporally consistent manner. Existing methods can match the human body with the clothing and then present it by the way of video, however it will cause the problem of excessive distortion of the grid and poor display effect at last. Given the problem, we found that was due to the neglect of the relationship between inputs lead to the loss of some features, while the conventional convolution operation is difficult to establish the remote information that is crucial in generating globally consistent results, restriction on clothing texture detail can lead to excessive deformation in the process of TPS fitting, make a lot of the try-on method in the final video rendering is not real. For the above problems, we reduce the excessive distortion of the garment during deformation by using a constraint function to regularize the TPS parameters; it also proposes a multimodal two-stage combinatorial Transformer: in the first stage, an interaction module is added, in which the long-distance relationship between people and clothing can be simulated, and then a better remote relationship can be obtained as well as contribute to the performance of TPS; in the second stage, an activation module is added, which can establish a global dependency, and this dependency can make the input important regions in the data are more prominent, which can provide more natural intermediate inputs for subsequent U-net networks. This paper’s method can bring better results for video virtual fitting, and experiments on the VVT dataset prove that the method outperforms previous methods in both quantitative and qualitative aspects.","PeriodicalId":111802,"journal":{"name":"Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry","volume":"222 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114595790","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Semantic-assisted Unified Network for Feature Point Extraction and Matching 语义辅助的特征点提取与匹配统一网络

Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry Pub Date : 2022-12-27 DOI: 10.1145/3574131.3574433

Daoming Ji, W. You, Yisong Chen, Guoping Wang, Sheng Li

{"title":"Semantic-assisted Unified Network for Feature Point Extraction and Matching","authors":"Daoming Ji, W. You, Yisong Chen, Guoping Wang, Sheng Li","doi":"10.1145/3574131.3574433","DOIUrl":"https://doi.org/10.1145/3574131.3574433","url":null,"abstract":"Feature point matching between two images is an essential part of 3D reconstruction, augmented reality, panorama stitching, etc. The quality of the initial feature point matching stage greatly affects the overall performance of a system. We present a unified feature point extraction-matching method, making use of semantic segmentation results to constrain feature point matching. To integrate high-level semantic information into feature points efficiently, we propose a unified feature point extraction and matching network, called SP-Net, which can detect feature points and generate feature descriptors simultaneously and perform feature point matching with accurate outcomes. Compared with previous works, our method can extract multi-scale context of the image, including shallow information and high-level semantic information of the local area, which is more stable when handling complex conditions such as changing illumination or large viewpoint. In evaluating the feature-matching benchmark, our method shows superior performance over the state-of-art method. As further validation, we propose SP-Net++ as an extension for 3D reconstruction. The experimental results show that our neural network can obtain accurate feature point positioning and robust feature matching to recover more cameras and get a well-shaped point cloud. Our semantic-assisted method can improve the stability of feature points as well as specific applicability for complex scenes.","PeriodicalId":111802,"journal":{"name":"Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125930052","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Exploring Text Selection in Augmented Reality Systems 探索增强现实系统中的文本选择

Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry Pub Date : 2022-12-27 DOI: 10.1145/3574131.3574459

Xinyi Liu, Xuanru Meng, Becky Spittle, Wenge Xu, Boyu Gao, Hai-Ning Liang

引用次数: 2

3D Grasp Pose Generation from 2D Anchors and Local Surface 从2D锚点和局部表面生成3D抓取姿势

Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry Pub Date : 2022-12-27 DOI: 10.1145/3574131.3574453

Hao Xu, Yangchang Sun, Qi Sun, Minghao Yang, Jinlong Chen, Baohua Qiang, Jinghong Wang

{"title":"3D Grasp Pose Generation from 2D Anchors and Local Surface","authors":"Hao Xu, Yangchang Sun, Qi Sun, Minghao Yang, Jinlong Chen, Baohua Qiang, Jinghong Wang","doi":"10.1145/3574131.3574453","DOIUrl":"https://doi.org/10.1145/3574131.3574453","url":null,"abstract":"This work proposes a three-dimensional (3D) robot grasp pose generation method for robot manipulators from the predicted two-dimensional (2D) anchors and the depth information of the local surface. Compared to the traditional image-based grasp area detection methods in which the grasp pose is only presented by two contact points, the proposed method can generate a more accurate 3D grasp pose. Furthermore, different from the 6-DoF object pose regression methods in which the point cloud of the whole objects is considered, the proposed method is very lightweight, since the 3D computation is only processed on the depth information of the local grasp surface. The method consists of three steps: (1) detecting the 2D grasp anchor and extracting the local grasp surface from the image; (2) obtaining the average vector of the objects’ local grasp surface from the objects’ local point cloud; (3) generating the 3D grasp pose from 2D grasp anchor based on the average vector of local grasp surface. The experiments are carried out on the Cornell and Jacquard grasp datasets. It is found that the proposed method yields improvement in the grasp accuracy compared to state-of-the-art 2D anchor methods. And the proposed method is also validated on the practical grasp tasks deployed on a UR5 arm with Robotiq Grippers F85. It outperforms state-of-the-art 2D anchor methods on the grasp success rate for dozens of practical grasp tasks.","PeriodicalId":111802,"journal":{"name":"Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129939060","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Exploring the Use of Smartphones as Input Devices for the Mixed Reality Environment 探索在混合现实环境中使用智能手机作为输入设备

Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry Pub Date : 2022-12-27 DOI: 10.1145/3574131.3574451

Zhigeng Pan, Zhipeng Pan, Tianren Luo, Mingmin Zhang

{"title":"Exploring the Use of Smartphones as Input Devices for the Mixed Reality Environment","authors":"Zhigeng Pan, Zhipeng Pan, Tianren Luo, Mingmin Zhang","doi":"10.1145/3574131.3574451","DOIUrl":"https://doi.org/10.1145/3574131.3574451","url":null,"abstract":"Nowadays, researches on mixed reality (MR) have made a lot of exploration in the aspects of user experience, hardware devices, interaction technologies, application systems, etc. However, more research is still needed to explore how to improve the experience in the shared MR Environment and design appropriate collaborative interaction models. The traditional VR handle can realize tool simulation and 3D interaction, while smartphones can realize 2D interactions such as fast 2D gestures, text input and handwriting. Using the mobile phone as the handle of the MR headset can realize the complementary advantages of the two. In this study, we design a cross-device collaborative system sharing a hybrid reality environment. In the MR Environment, the smartphone will realize the functions of the controller for remote selection and manipulation of objects, and its advantages in 2D interaction can be well applied to some special tasks. We design the user interface with three levels of immersion: high, medium and low. High immersion: We provide a hybrid user interface combining a HoloLens2 and a smartphone. Medium immersion: We provide a 3D user interface represented by HoloLens2. Low immersion: We provide a multi-touch user interface represented by tablet. User studies show that the hybrid user interfaces can bring users satisfactory immersion and interactive experience, but it also needs to design accurate and efficient input methods according to the interactive tasks.","PeriodicalId":111802,"journal":{"name":"Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125305106","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Simulation of resin filling distance during the pulling-up stage for rapid DLP printing 快速DLP打印拉升阶段树脂填充距离的模拟

Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry Pub Date : 2022-12-27 DOI: 10.1145/3574131.3574437

Li-dong Zhao, Zening Men, Lifang Wu, Zhifeng Zhao, Yan Zhang, Ge Shi

{"title":"Simulation of resin filling distance during the pulling-up stage for rapid DLP printing","authors":"Li-dong Zhao, Zening Men, Lifang Wu, Zhifeng Zhao, Yan Zhang, Ge Shi","doi":"10.1145/3574131.3574437","DOIUrl":"https://doi.org/10.1145/3574131.3574437","url":null,"abstract":"In continuous digital light processing (DLP) 3D printing, the continuous lifting in the pulling-up stage may result in an insufficient resin filling for large-area structures, which limits its applications. Some researchers proposed a concept of maximum filling distance (MFD) and combined continuous and layer-wise printing for large-area rapid printing (LARP). However, the current MFD is estimated by experiments, which is far from optimal. In this paper, a computational fluid dynamics (CFD) analysis method is proposed to simulate the resin filling during the pulling-up stage for rapid DLP printing. Poiseuille flow is used to establish the CFD analysis models. The Volume of Fluid method is designed to simulate the resin filling behavior. The MFD can be predicted from the simulations. The validation of predicted results are fulfilled by actual printing processes. Printing experiments show that the proposed method can be successfully applied to different resins and printing parameters. It increases the efficiency of printing a new model or developing novel printing materials.","PeriodicalId":111802,"journal":{"name":"Proceedings of the 18th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123037876","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0