2020 International Conference on 3D Immersion (IC3D)最新文献

A Novel 3D-Unet Deep Learning Framework Based on High-Dimensional Bilateral Grid for Edge Consistent Single Image Depth Estimation 基于高维双边网格的3D-Unet深度学习框架边缘一致单幅图像深度估计

2020 International Conference on 3D Immersion (IC3D) Pub Date : 2020-12-15 DOI: 10.1109/IC3D51119.2020.9376327

Mansi Sharma, Abheesht Sharma, Kadvekar Rohit Tushar, Avinash Panneer

{"title":"A Novel 3D-Unet Deep Learning Framework Based on High-Dimensional Bilateral Grid for Edge Consistent Single Image Depth Estimation","authors":"Mansi Sharma, Abheesht Sharma, Kadvekar Rohit Tushar, Avinash Panneer","doi":"10.1109/IC3D51119.2020.9376327","DOIUrl":"https://doi.org/10.1109/IC3D51119.2020.9376327","url":null,"abstract":"The task of predicting smooth and edge-consistent depth maps is notoriously difficult for single image depth estimation. This paper proposes a novel Bilateral Grid based 3D convolutional neural network, dubbed as 3DBG-UNet, that parameterize high dimensional feature space by encoding compact 3D bilateral grids with UNets and infers sharp geometric layout of the scene. Further, an another novel 3DBGES-UNet model is introduced that integrate 3DBG-UNet for inferring an accurate depth map given a single color view. The 3DBGES-UNet concatenate 3DBG-UNet geometry map with the inception network edge accentuation map and a spatial object's boundary map obtained by leveraging semantic segmentation and train the UNet model with ResNet backbone. Both models are designed with a particular attention to explicitly account for edges or minute details. Preserving sharp discontinuities at depth edges is critical for many applications such as realistic integration of virtual objects in AR video or occlusion-aware view synthesis for 3D display applications. The proposed depth prediction network achieves state-of-the-art performance in both qualitative and quantitative evaluations on the challenging NYUv2-Depth data. The code and corresponding pre-trained weights will be made publicly available.","PeriodicalId":159318,"journal":{"name":"2020 International Conference on 3D Immersion (IC3D)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129458331","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: 5

[Copyright notice] (版权)

2020 International Conference on 3D Immersion (IC3D) Pub Date : 2020-12-15 DOI: 10.1109/ic3d51119.2020.9376333

引用次数: 0

A Device for Capturing Inward-Looking Spherical Light Fields 一种捕捉内向球形光场的装置

2020 International Conference on 3D Immersion (IC3D) Pub Date : 2020-12-15 DOI: 10.1109/IC3D51119.2020.9376346

Quentin Bolsée, W. Darwish, Daniele Bonatto, G. Lafruit, A. Munteanu

引用次数: 0

Synthesis of Computer Generated Holograms Using Layer-Based Method and Perspective Projection Images 基于层的方法合成计算机生成全息图和透视投影图像

2020 International Conference on 3D Immersion (IC3D) Pub Date : 2020-12-15 DOI: 10.1109/IC3D51119.2020.9376326

Vincent Brac de la Perriere, V. Drazic, D. Doyen, Arno Schubert

{"title":"Synthesis of Computer Generated Holograms Using Layer-Based Method and Perspective Projection Images","authors":"Vincent Brac de la Perriere, V. Drazic, D. Doyen, Arno Schubert","doi":"10.1109/IC3D51119.2020.9376326","DOIUrl":"https://doi.org/10.1109/IC3D51119.2020.9376326","url":null,"abstract":"From its Nobel prize winning discovery by Denis Gabor over half a century ago, Holography has been alternately put in the spotlight as a promising technique for its capacity of displaying 3D scenes, to be later forgotten regarding the complexity of performing holographic recording outside from optical laboratories. The later development of high resolution microdis-plays, high capacity CPU and GPU units raised the interest of using Holography as a mean to reproduce dynamic 3D scenes through the computation of holograms and the display of the associated interference pattern on a Spatial Light Modulators (SLMs). The computation process can for example be performed from a layered Computer Generated (CG) scene, using well-known convolution methods or other transfer function models from Fourier optics. The simplicity and speed of this technique is unfortunately compensated by its limitation to small object and scenes. We here want to demonstrate the possibility to use similar techniques to compute holograms of larger field of view, using perspective-projection layered scene and well-known Fourier optics tools.","PeriodicalId":159318,"journal":{"name":"2020 International Conference on 3D Immersion (IC3D)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121217071","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

“The Mystery of the Raddlesham Mumps”: a Case Study for Combined Storytelling in a Theatre Play and Virtual Reality 《拉德尔沙姆流行性腮腺炎之谜》:戏剧与虚拟现实结合叙事的案例研究

2020 International Conference on 3D Immersion (IC3D) Pub Date : 2020-12-15 DOI: 10.1109/IC3D51119.2020.9376391

L. Dam, Abigail L. M. Webb, Liam Jarvis, P. Hibbard, M. Linley

{"title":"“The Mystery of the Raddlesham Mumps”: a Case Study for Combined Storytelling in a Theatre Play and Virtual Reality","authors":"L. Dam, Abigail L. M. Webb, Liam Jarvis, P. Hibbard, M. Linley","doi":"10.1109/IC3D51119.2020.9376391","DOIUrl":"https://doi.org/10.1109/IC3D51119.2020.9376391","url":null,"abstract":"“The Mystery of the Raddlesham Mumps” is a poem by Murray Lachlan Young, aimed at both children and adults. This poem has been adapted as a theatre play with a short prequel as a Virtual Reality (VR) / tablet app. We used this unique combination to explore the potential interaction between these different media elements for the level of “pres-ence” and “immersion” in the story (i.e. the level to which one can imagine oneself within the story at the expense of the sense of physical time and space). The theatre audience had the opportunity to play the VR / tablet app in the foyer before the performance started. After the performance, a questionnaire measured participants' level of immersion and presence in the theatre play and their enjoyment of both play and app. The results showed that people of all ages interacted with and liked the app. Ratings for the play were also high and did not depend on prior engagement with the app. However, the play was liked more by adults than children, and the reverse was true for the app, suggesting a potential generation shift in multimedia story telling.","PeriodicalId":159318,"journal":{"name":"2020 International Conference on 3D Immersion (IC3D)","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131665591","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

Neural-Network-Based Detection Methods for Color, Sharpness, and Geometry Artifacts in Stereoscopic and VR180 Videos 基于神经网络的立体和VR180视频中颜色、清晰度和几何伪影的检测方法

2020 International Conference on 3D Immersion (IC3D) Pub Date : 2020-12-15 DOI: 10.1109/IC3D51119.2020.9376385

S. Lavrushkin, Konstantin Kozhemyakov, D. Vatolin

引用次数: 0

A Practical Approach for Microphone Array Calibration in Augmented and Virtual Reality Applications 增强现实和虚拟现实应用中麦克风阵列标定的实用方法

2020 International Conference on 3D Immersion (IC3D) Pub Date : 2020-12-15 DOI: 10.1109/IC3D51119.2020.9376386

Noman Akbar, G. Dickins, Mark R. P. Thomas

引用次数: 0

Robust Calibration of a Multi-View Azure Kinect Scanner Based on Spatial Consistency 基于空间一致性的多视图Azure Kinect扫描仪鲁棒校准

2020 International Conference on 3D Immersion (IC3D) Pub Date : 2020-12-15 DOI: 10.1109/IC3D51119.2020.9376321

W. Darwish, Quentin Bolsée, A. Munteanu

引用次数: 1

Training Firefighters in Virtual Reality 在虚拟现实中训练消防员

2020 International Conference on 3D Immersion (IC3D) Pub Date : 2020-12-15 DOI: 10.1109/IC3D51119.2020.9376336

Michel Bellemans, D. Lamrnens, J. D. Sloover, Tom De Vleeschauwer, Evarest Schoofs, W. Jordens, B. V. Steenhuyse, J. Mangelschots, S. Selleri, Charles Hamesse, Timothée Fréville, R. Haeltermani

引用次数: 4

A Unified Deep Learning Approach for Foveated Rendering & Novel View Synthesis from Sparse RGB-D Light Fields 基于稀疏RGB-D光场的注视点渲染和新颖视图合成的统一深度学习方法

2020 International Conference on 3D Immersion (IC3D) Pub Date : 2020-12-15 DOI: 10.1109/IC3D51119.2020.9376340

Vineet Thumuluri, Mansi Sharma

{"title":"A Unified Deep Learning Approach for Foveated Rendering & Novel View Synthesis from Sparse RGB-D Light Fields","authors":"Vineet Thumuluri, Mansi Sharma","doi":"10.1109/IC3D51119.2020.9376340","DOIUrl":"https://doi.org/10.1109/IC3D51119.2020.9376340","url":null,"abstract":"Near-eye light field displays provide a solution to visual discomfort when using head mounted displays by presenting accurate depth and focal cues. However, light field HMDs require rendering the scene from a large number of viewpoints. This computational challenge of rendering sharp imagery of the foveal region and reproduce retinal defocus blur that correctly drives accommodation is tackled in this paper. We designed a novel end-to-end convolutional neural network that leverages human vision to perform both foveated reconstruction and view synthesis using only 1.2% of the total light field data. The proposed architecture comprises of log-polar sampling scheme followed by an interpolation stage and a convolutional neural network. To the best of our knowledge, this is the first attempt that synthesizes the entire light field from sparse RGB-D inputs and simultaneously addresses foveation rendering for computational displays. Our algorithm achieves fidelity in the fovea without any perceptible artifacts in the peripheral regions. The performance in fovea is comparable to the state-of-the-art view synthesis methods, despite using around 10x less light field data.","PeriodicalId":159318,"journal":{"name":"2020 International Conference on 3D Immersion (IC3D)","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131511159","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: 4