6,122 Hits in 3.8 sec

Completed Dense Scene Flow in RGB-D Space [chapter]

Yucheng Wang, Jian Zhang, Zicheng Liu, Qiang Wu, Philip Chou, Zhengyou Zhang, Yunde Jia
2015 Lecture Notes in Computer Science  
In this paper, a completed dense scene flow framework is proposed, which models both rotation and translation for general motion estimation.  ...  The proposed framework is applied on the RGB-D image space where the computation efficiency is further improved.  ...  We also acknowledge Minqi Li for recording the Kinect RGB-D data sequence in our experiment.  ... 
doi:10.1007/978-3-319-16628-5_14 fatcat:x5ml6cofnfgr7bf4rpwvjzvndq

High-quality indoor scene 3D reconstruction with RGB-D cameras: A brief review

Jianwei Li, Wei Gao, Yihong Wu, Yangdong Liu, Yanfei Shen
2022 Computational Visual Media  
The advent of consumer RGB-D cameras has made a profound advance in indoor scene reconstruction.  ...  We here review high-quality 3D indoor scene reconstruction methods using consumer RGB-D cameras.  ...  The pre-processed RGB-D images with estimated camera poses are integrated into a complete 3D scene model.  ... 
doi:10.1007/s41095-021-0250-8 fatcat:z6ywcn4zujbptjaqrwkwhoyquu

3D+T dense motion trajectories as kinematics primitives to recognize gestures on depth video sequences

Fabio Martínez Carrillo, Fabián Castillo, Lola Bautista
2019 Revista Politécnica  
Regarding motion characterization, typical RGB-D strategies are limited to namely analyze global shape changes and capture scene flow fields to describe local motions in depth sequences.  ...  RGB-D sensors have allowed attacking many classical problems in computer vision such as segmentation, scene representations and human interaction, among many others.  ...  Inspired in such dense trajectories, in this work is proposed an extension of dense trajectories computed in RGB-D spaces, which allows capturing Then, between consecutive frames is computed a scene flow  ... 
doi:10.33571/rpolitec.v15n29a7 fatcat:dk5g4laa2rc7vct3ea75psd4kq

Action Recognition from RGB-D Data: Comparison and Fusion of Spatio-Temporal Handcrafted Features and Deep Strategies

Maryam Asadi-Aghbolaghi, Hugo Bertiche, Vicent Roig, Shohreh Kasaei, Sergio Escalera
2017 2017 IEEE International Conference on Computer Vision Workshops (ICCVW)  
In this research, Multimodal dense trajectories (MMDT) is proposed to describe RGB-D videos. Dense trajectories are pruned based on scene flow data.  ...  In this work, multimodal fusion of RGB-D data are analyzed for action recognition by using scene flow as early fusion and integrating the results of all modalities in a late fusion fashion.  ...  Jaimez et al. in [9] proposed the first dense real-time scene flow algorithm for RGB-D cameras.  ... 
doi:10.1109/iccvw.2017.376 dblp:conf/iccvw/Asadi-Aghbolaghi17 fatcat:cv3r4plnnffjlgxs3dpqmlo5xy

Automatic layered RGB-D scene flow estimation with optical flow field constraint

Xiuxiu Li, Yanjuan Liu, Haiyan Jin, Jiangbin Zheng, Lei Cai
2020 IET Image Processing  
Scene flow estimation with RGB-D frames is receiving increasing attention in digital video processing and computer vision due to the widespread use of depth sensors.  ...  To this end, an automatic layered RGB-D scene flow estimation method is proposed, which achieves more accurate layering of objects in depth image by exploring motion information.  ...  [14] presented a dense real-time scene flow algorithm with brightness constancy and geometric consistency for RGB-D camera.  ... 
doi:10.1049/iet-ipr.2020.0230 fatcat:7u66iqzczvfx5o6e6gpydjg3n4

Dynamic Dense RGB-D SLAM using Learning-based Visual Odometry [article]

Shihao Shen, Yilin Cai, Jiayi Qiu, Guangzhao Li
2022 arXiv   pre-print
We propose a dense dynamic RGB-D SLAM pipeline based on a learning-based visual odometry, TartanVO.  ...  TartanVO, like other direct methods rather than feature-based, estimates camera pose through dense optical flow, which only applies to static scenes and disregards dynamic objects.  ...  The architecture of our dense RGB-D SLAM pipeline using 2D scene flow based segmentation is shown in Fig. 2 .  ... 
arXiv:2205.05916v1 fatcat:p6xojvqeebcanl3mmigrngwoza

Real-Time Dense Reconstruction of Indoor Scene

Jinxing Niu, Qingsheng Hu, Yi Niu, Tao Zhang, Sunil Kumar Jha
2021 Computers Materials & Continua  
A Realsense D415 sensor is used as RGB-D cameras to obtain the three-dimensional (3D) point clouds of an indoor environments.  ...  A parallel thread of map reconstruction is added with surfel model, and depth map and RGB map are fused to build the dense map.  ...  The surfel map is used to reconstruct the dense indoor scene. The improved ORB-SLAM2 flow diagram is shown as Fig. 6 , the red line frame in the figure is the improved parts.  ... 
doi:10.32604/cmc.2021.017418 fatcat:j7nv25g7pncbrixk3u3zulhe3e

DeepLiDARFlow: A Deep Learning Architecture For Scene Flow Estimation Using Monocular Camera and Sparse LiDAR [article]

Rishav, Ramy Battrawy, René Schuster, Oliver Wasenmüller, Didier Stricker
2020 arXiv   pre-print
In this paper, we present DeepLiDARFlow, a novel deep learning architecture which fuses high level RGB and LiDAR features at multiple scales in a monocular setup to predict dense scene flow.  ...  Scene flow is the dense 3D reconstruction of motion and geometry of a scene. Most state-of-the-art methods use a pair of stereo images as input for full scene reconstruction.  ...  ACKNOWLEDGMENT This work was partially funded by the Federal Ministry of Education and Research Germany in the project VIDETE (01IW18002).  ... 
arXiv:2008.08136v1 fatcat:dhypelqe4ncetfvj5fxmgr6vwy

Speed and Memory Efficient Dense RGB-D SLAM in Dynamic Scenes

Bruce Canovas, Michele Rombaut, Amaury Negre, Denis Pellerin, Serge Olympieff
2020 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)  
In this paper we propose a fast RGB-D SLAM articulated around a rough and lightweight 3D representation for dense compact mapping in dynamic indoor environment, targeting mainstream computing platforms  ...  Real-time dense 3D localization and mapping systems are required to enable robotics platforms to interact in and with their environments.  ...  CONCLUSION In this paper we described a nnew complete dense RGB-D SLAM pipeline.  ... 
doi:10.1109/iros45743.2020.9341542 fatcat:na7wddpekzd6jei4zfbyta53yy

Onboard dynamic RGB‐D simultaneous localization and mapping for mobile robot navigation

Bruce Canovas, Amaury Nègre, Michèle Rombaut
2021 ETRI Journal  
To mitigate these issues, we propose a completely closed-loop online dense RGB-D SLAM algorithm targeting autonomous indoor mobile robot navigation tasks.  ...  dynamics in RGB-D point clouds.  ... 
doi:10.4218/etrij.2021-0061 fatcat:533437je5bfrjigf3qxvbusbfi

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
Unfortunately, in the context of RGB-D scene understanding, very little data is available -- current datasets cover a small range of scene views and have limited semantic annotations.  ...  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

DeepDeform: Learning Non-rigid RGB-D Reconstruction with Semi-supervised Data [article]

Aljaž Božič, Michael Zollhöfer, Christian Theobalt, Matthias Nießner
2020 arXiv   pre-print
This way, we obtain a large dataset of 400 scenes, over 390,000 RGB-D frames, and 5,533 densely aligned frame pairs; in addition, we provide a test set along with several metrics for evaluation.  ...  Here, we propose a new neural network that operates on RGB-D frames, while maintaining robustness under large non-rigid deformations and producing accurate predictions.  ...  Acknowledgements We would like to thank the expert annotators Sathya Ashok, Omar Hedayat, Haya Irfan, Azza Jenane, Soh Yee Lee, Suzana Spasova, and Weile Weng for their efforts in building the dataset.  ... 
arXiv:1912.04302v2 fatcat:7t2d3xekyvey3doiszgnjqkwe4

Activity recognition in unconstrained RGB-D video using 3D trajectories

Yang Xiao, Gangqiang Zhao, Junsong Yuan, Daniel Thalmann
2014 SIGGRAPH Asia 2014 Autonomous Virtual Humans and Social Robot for Telepresence on - SIGGRAPH ASIA '14  
We develop a novel RGB-D activity recognition approach that leverages the dense trajectory feature in RGB videos.  ...  Human activity recognition in unconstrained RGB-D videos has extensive applications in surveillance, multimedia data analytics, human-computer interaction, etc, but remains a challenging problem due to  ...  Most of this work was done at Nanyang Technological University, when Yang Xiao worked as the research fellow in IMI under the supervision of Prof. Junsong Yuan and Prof. Daniel Thalmann.  ... 
doi:10.1145/2668956.2668961 dblp:conf/siggraph/XiaoZYT14 fatcat:gtyopbq5izf3ldb2xvwjh6e7ja

3dDepthNet: Point Cloud Guided Depth Completion Network for Sparse Depth and Single Color Image [article]

Rui Xiang, Feng Zheng, Huapeng Su, Zhe Zhang
2020 arXiv   pre-print
the original RGB-D images to perform 2D image completion.  ...  Depth densification is first performed in 3D space via point cloud completion, followed by a specially designed encoder-decoder structure that utilizes the projected dense depth from 3D completion and  ...  RGB-D camera, are usually more accurate because of the limited distance of indoor scenes and good illumination conditions [13, 38] .  ... 
arXiv:2003.09175v1 fatcat:lolswhln6rftxfvzjgm6azutae

3D Semantic Scene Completion: a Survey [article]

Luis Roldao, Raoul de Charette, Anne Verroust-Blondet
2021 arXiv   pre-print
Semantic Scene Completion (SSC) aims to jointly estimate the complete geometry and semantics of a scene, assuming partial sparse input.  ...  Specifically, SSC lies in the ambiguous completion of large unobserved areas and the weak supervision signal of the ground truth.  ...  The same network is used in [33] to predict complete scenes from panoramic RGB-D images.  ... 
arXiv:2103.07466v3 fatcat:swz4azlznre3laziatls6sdrfm
« Previous Showing results 1 — 15 out of 6,122 results