Filters








4,235 Hits in 5.1 sec

Environment reconstruction on depth images using Generative Adversarial Networks [article]

Lucas P. N. Matias, Jefferson R. Souza, Denis F. Wolf
2019 arXiv   pre-print
We then introduce a loss function focused on disparity and environment depth data reconstruction, and a Generative Adversarial Network (GAN) architecture able to deal with occluded information inference  ...  Our final contribution is a loss function focused on disparity data and a GAN able to extract depth features and estimate depth data by inpainting disparity images.  ...  The Surface Attention branch improves the structures reconstruction by propagating real surface information to the generated area.  ... 
arXiv:1912.03992v1 fatcat:27s4w2y4ljfe5k43bsdzkio7me

FusionMLS: Highly dynamic 3D reconstruction with consumer-grade RGB-D cameras

Siim Meerits, Diego Thomas, Vincent Nozick, Hideo Saito
2018 Computational Visual Media  
In this paper, we demonstrate that multiple unsynchronized low-cost RGB-D cameras can be used for the same purpose.  ...  Our approach does not store the reconstruction volume in memory, making it memory-efficient and scalable to large scenes. Our implementation is completely GPU based and works in real time.  ...  Figure 4 shows an example of interpolation at three different time points using real data.  ... 
doi:10.1007/s41095-018-0121-0 fatcat:uiqymh55ibfmjczikwike5lpma

ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes [article]

Angela Dai, Angel X. Chang, Manolis Savva, Maciej Halber, Thomas Funkhouser, Matthias Nießner
2017 arXiv   pre-print
To collect this data, we designed an easy-to-use and scalable RGB-D capture system that includes automated surface reconstruction and crowdsourced semantic annotation.  ...  To address this issue, we introduce ScanNet, an RGB-D video dataset containing 2.5M views in 1513 scenes annotated with 3D camera poses, surface reconstructions, and semantic segmentations.  ...  Last but not least, we would like to thank all the volunteers who helped with scanning and get-ting us access to scanning spaces.  ... 
arXiv:1702.04405v2 fatcat:nxt2fhgtsfdmlih7ukyoedoc4q

Live User-Guided Intrinsic Video for Static Scenes

Abhimitra Meka, Gereon Fox, Michael Zollhofer, Christian Richardt, Christian Theobalt
2017 IEEE Transactions on Visualization and Computer Graphics  
Abstract-We present a novel real-time approach for user-guided intrinsic decomposition of static scenes captured by an RGB-D sensor.  ...  We propose a novel approach for live, user-guided intrinsic video decomposition. We first obtain a dense volumetric reconstruction of the scene using a commodity RGB-D sensor.  ...  ACKNOWLEDGMENTS We thank the anonymous reviewers for their helpful feedback. This work was supported by the ERC Starting Grant CapReal (335545).  ... 
doi:10.1109/tvcg.2017.2734425 pmid:28809688 fatcat:2ieprbfyt5cqdaq57ywy5ngrc4

Shading-based refinement on volumetric signed distance functions

Michael Zollhöfer, Angela Dai, Matthias Innmann, Chenglei Wu, Marc Stamminger, Christian Theobalt, Matthias Nießner
2015 ACM Transactions on Graphics  
In our approach, we leverage RGB data to refine these reconstructions through shading cues, as color input is typically of much higher resolution than the depth data.  ...  Exploiting the shading cues from the RGB data allows us to obtain reconstructions at previously unseen resolutions within only a few seconds.  ...  Acknowledgements We would like to thank Qian-Yi Zhou and Vladlen Koltun for the Fountain data.  ... 
doi:10.1145/2766887 fatcat:yflg5nbvazct7hz224pynzkkdi

Dense tracking, mapping and scene labeling using a depth camera

Andrés Alejandro Díaz-Toro, Lina María Paz-Pérez, Pedro Piniés-Rodríguez, Eduardo Francisco Caicedo-Bravo
2018 Revista Facultad de Ingeniería  
tracking, 3D reconstruction, and object detection of desktop-like environments, using a depth camera; the Kinect sensor.  ...  Alternatively, the user can couple the object detection module (YOLO: you only look once [1]) for detecting and propagating to the model information of categories of objects commonly found over desktops  ...  The authors would like to thank Fundación CEIBA for the financial support that has made the development of this work possible.  ... 
doi:10.17533/udea.redin.n86a07 fatcat:sda3oc6pmvaexbgjr3ooy5o6cq

Real-time and scalable incremental segmentation on dense SLAM

Keisuke Tateno, Federico Tombari, Nassir Navab
2015 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)  
This work proposes a real-time segmentation method for 3D point clouds obtained via Simultaneous Localization And Mapping (SLAM).  ...  The proposed method incrementally merges segments obtained from each input depth image in a unified global model using a SLAM framework.  ...  For the comparison, we have used one sequence (fr1/room) from the public TUM RGB-D SLAM benchmark dataset [22] .  ... 
doi:10.1109/iros.2015.7354011 dblp:conf/iros/TatenoTN15 fatcat:z7kqcrc6tfghtcivoymlj55b2a

VolumeDeform: Real-time Volumetric Non-rigid Reconstruction [article]

Matthias Innmann, Michael Zollhöfer, Matthias Nießner, Christian Theobalt, Marc Stamminger
2016 arXiv   pre-print
We present a novel approach for the reconstruction of dynamic geometric shapes using a single hand-held consumer-grade RGB-D sensor at real-time rates.  ...  The problem is tackled in real-time at the camera's capture rate using a data-parallel flip-flop optimization strategy.  ...  Acknowledgments We thank Angela Dai for the video voice over and Richard Newcombe for the Dy-namicFusion comparison sequences.  ... 
arXiv:1603.08161v2 fatcat:sksadhn6rrdyhcvq2ve4osx2d4

STAR3D: Simultaneous Tracking and Reconstruction of 3D Objects Using RGB-D Data

Carl Yuheng Ren, Victor Prisacariu, David Murray, Ian Reid
2013 2013 IEEE International Conference on Computer Vision  
Surface and background appearance models are served depth/RGB-D data.  ...  using RGB-D data.  ... 
doi:10.1109/iccv.2013.197 dblp:conf/iccv/RenPMR13 fatcat:3dywhvyb4zg5tg2afkxzvjfgw4

VolumeDeform: Real-Time Volumetric Non-rigid Reconstruction [chapter]

Matthias Innmann, Michael Zollhöfer, Matthias Nießner, Christian Theobalt, Marc Stamminger
2016 Lecture Notes in Computer Science  
We present a novel approach for the reconstruction of dynamic geometric shapes using a single hand-held consumer-grade RGB-D sensor at real-time rates.  ...  The problem is tackled in real-time at the camera's capture rate using a data-parallel flip-flop optimization strategy.  ...  Acknowledgments We thank Angela Dai for the video voice over and Richard Newcombe for the Dy-namicFusion comparison sequences.  ... 
doi:10.1007/978-3-319-46484-8_22 fatcat:sdmmq5ukmffordivhxkvx3j7e4

MonoFusion: Real-time 3D reconstruction of small scenes with a single web camera

Vivek Pradeep, Christoph Rhemann, Shahram Izadi, Christopher Zach, Michael Bleyer, Steven Bathiche
2013 2013 IEEE International Symposium on Mixed and Augmented Reality (ISMAR)  
Figure 1 : We present a new system for real-time 3D reconstruction using a single moving off-the-shelf web camera. Our system first estimates the pose of the camera using a sparse feature tracker.  ...  This allows cheap and readily-available web cameras to be used for small sized reconstructions for AR applications.  ...  Figures 3 and 1 show how small scenes can be reconstructed in real-time and finally texture mapped using the RGB data.  ... 
doi:10.1109/ismar.2013.6671767 dblp:conf/ismar/PradeepRIZBB13 fatcat:j7slcu3s5zcy7ovfukqpzqza4e

Online Reconstruction of Indoor Scenes from RGB-D Streams

Hao Wang, Jun Wang, Liang Wang
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
A system capable of performing robust online volumetric reconstruction of indoor scenes based on input from a handheld RGB-D camera is presented.  ...  The tracker operates in real-time, which allows the reconstruction results to be visualized during the scanning process. Live visual feedbacks makes the scanning operation fast and intuitive.  ...  Our camera tracking via volumetric Model Quality Evaluation For RGB-D dense reconstruction, based on our knowledge, there is no ground truth on real data for evaluating reconstruction accuracy.  ... 
doi:10.1109/cvpr.2016.356 dblp:conf/cvpr/WangWW16 fatcat:bpokcmaapff3dkw7cjylkc5gce

Multi-Resolution 3D Mapping with Explicit Free Space Representation for Fast and Accurate Mobile Robot Motion Planning [article]

Nils Funk, Juan Tarrio, Sotiris Papatheodorou, Marija Popovic, Pablo F. Alcantarilla, Stefan Leutenegger
2021 arXiv   pre-print
We introduce a method for choosing resolution -- on the fly -- in real-time by means of a multi-scale max-min pooling of the input depth image.  ...  The notion of explicit free space mapping paired with the spatial hierarchy in the data structure, as well as map resolution, allows for collision queries, as needed for robot motion planning, at unprecedented  ...  )): quadratic for RGB-D and linear for LIDAR or synthetic (perfect) depth cameras.  ... 
arXiv:2010.07929v5 fatcat:as2eaqmg6zgg7j46zxmklzm2v4

Guest Editors' Introduction to the Special Issue on RGB-D Vision: Methods and Applications

Mohammed Bennamoun, Yulan Guo, Federico Tombari, Kamal Youcef-Toumi, Ko Nishino
2020 IEEE Transactions on Pattern Analysis and Machine Intelligence  
Compared to 2D images and 3D data (including depth images, point clouds and meshes), RGB-D images represent both the photometric and geometric information of a scene.  ...  In the last few years, a large number of RGB-D datasets have also been publicly released to tackle various vision tasks.  ...  Liu proposes a new real-time system that combines volumetric non-rigid reconstruction with data-driven template fitting to simultaneously reconstruct detailed surface geometry, large non-rigid motion,  ... 
doi:10.1109/tpami.2020.2976227 fatcat:dqt5dt3ymnesfikmgu2sxffdcu

NRMVS: Non-Rigid Multi-View Stereo [article]

Matthias Innmann, Kihwan Kim, Jinwei Gu, Matthias Niessner, Charles Loop, Marc Stamminger, Jan Kautz
2019 arXiv   pre-print
Multi-view Stereo (MVS) is a common solution in photogrammetry applications for the dense reconstruction of a static scene.  ...  Scene reconstruction from unorganized RGB images is an important task in many computer vision applications.  ...  Figure 6 :Figure 7 : 67 Qualitative evaluation with real data: In each row, the first two columns show the views used to create each canonical surface.  ... 
arXiv:1901.03910v1 fatcat:tg4vchqqyjhsjlebnk6qwuw2w4
« Previous Showing results 1 — 15 out of 4,235 results