Filters








1,389 Hits in 7.5 sec

Joint stereo 3D object detection and implicit surface reconstruction [article]

Shichao Li, Kwang-Ting Cheng
2022 arXiv   pre-print
We present the first learning-based framework for category-level 3D object detection and implicit shape estimation based on a pair of stereo RGB images in the wild.  ...  Previous stereo 3D object detection approaches cannot describe the complete shape details of the detected objects and often fails for the small objects.  ...  CONCLUSION We propose the first approach for joint stereo 3D object detection and implicit shape reconstruction with a new twostage model S-3D-RCNN.  ... 
arXiv:2111.12924v2 fatcat:rtongqxoefd7vg7i3dsqm6dbua

DSP-SLAM: Object Oriented SLAM with Deep Shape Priors [article]

Jingwen Wang, Martin Rünz, Lourdes Agapito
2021 arXiv   pre-print
DSP-SLAM takes as input the 3D point cloud reconstructed by a feature-based SLAM system and equips it with the ability to enhance its sparse map with dense reconstructions of detected objects.  ...  We propose DSP-SLAM, an object-oriented SLAM system that builds a rich and accurate joint map of dense 3D models for foreground objects, and sparse landmark points to represent the background.  ...  We thank Wonbong Jang and Adam Sherwood for fruitful discussions.  ... 
arXiv:2108.09481v2 fatcat:2cxcoerz6vfjnkju5oskhxxuie

DSP-SLAM: Object Oriented SLAM with Deep Shape Priors

Jingwen Wang, Martin Runz, Lourdes Agapito
2021 2021 International Conference on 3D Vision (3DV)  
Reconstructed map and camera trajectory on KITTI 00.  ...  Figure 1: DSP-SLAM builds a rich object-aware map, providing complete detailed shapes of detected objects, while representing the background coarsely as sparse feature points.  ...  We thank Wonbong Jang and Adam Sherwood for fruitful discussions.  ... 
doi:10.1109/3dv53792.2021.00143 fatcat:zle434gjcvaatklyfrt7dyhzze

MonoPerfCap: Human Performance Capture from Monocular Video [article]

Weipeng Xu, Avishek Chatterjee, Michael Zollhöfer, Helge Rhodin, Dushyant Mehta, Hans-Peter Seidel, Christian Theobalt
2018 arXiv   pre-print
We tackle these challenges by a novel approach that employs sparse 2D and 3D human pose detections from a convolutional neural network using a batch-based pose estimation strategy.  ...  Our approach reconstructs articulated human skeleton motion as well as medium-scale non-rigid surface deformations in general scenes.  ...  In order to prune frames with low 3D detection confidence, we measure the per-frame PCK error [Toshev and Szegedy 2014] PCK f between the 2D joint detections and the projected 3D detections and apply  ... 
arXiv:1708.02136v2 fatcat:toewmmbynnbppmxsop43d4xk3e

Reconstruction for 3D immersive virtual environments

D. S. Alexiadis, G. Kordelas, K. C Apostolakis, J. D. Agapito, J. M. Vegas, E. Izquierdo, P. Daras
2012 2012 13th International Workshop on Image Analysis for Multimedia Interactive Services  
The paper focuses on techniques for the real-time, 3D reconstruction of moving humans from multiple Kinect devices.  ...  The future of tele-conferencing is towards multi-party 3D Tele-Immersion (TI) and TI environments that can support realistic inter-personal communications and virtual interaction among participants.  ...  (b) Reconstructed dense mesh from 4 stereo pairs. Fig. 7 . 7 Animated dancing avatar and a reconstructed human.  ... 
doi:10.1109/wiamis.2012.6226760 dblp:conf/wiamis/AlexiadisKAAVID12 fatcat:6h6hu54ywvbufk6dtv4ahpm2hu

Grid-Based Active Stereo with Single-Colored Wave Pattern for Dense One-shot 3D Scan

Ryusuke Sagawa, Kazuhiro Sakashita, Nozomu Kasuya, Hiroshi Kawasaki, Ryo Furukawa, Yasushi Yagi
2012 2012 Second International Conference on 3D Imaging, Modeling, Processing, Visualization & Transmission  
To achieve these goals, we propose the following methods: 1) implicit encoding of projector information by a grid of wave lines, 2) grid-based stereo between projector pattern and camera images to determine  ...  In this paper, we propose a method to reconstruct the shapes of moving objects.  ...  Acknowledgment This work was supported in part by SCOPE No.101710002 and NEXT program No.LR030 in Japan.  ... 
doi:10.1109/3dimpvt.2012.41 dblp:conf/3dim/SagawaSKKFY12 fatcat:t7zzr5laurbl5gvne52yw5bisu

3DFS: Deformable Dense Depth Fusion and Segmentation for Object Reconstruction from a Handheld Camera [article]

Tanmay Gupta, Daeyun Shin, Naren Sivagnanadasan, Derek Hoiem
2016 arXiv   pre-print
We propose an approach for 3D reconstruction and segmentation of a single object placed on a flat surface from an input video.  ...  We evaluate 3D reconstructions qualitatively on our Object-Videos dataset, comparing to fusion, multiview stereo, and segmentation baselines.  ...  We are also thankful to David Forsyth for helpful discussion on linearized bending energy and smoothness priors, and to Jason Rock for suggesting the region of interest based component of superpixel unary  ... 
arXiv:1606.05002v2 fatcat:pgoj22lywbfbzmmyo5i6gofqwu

IMPLICITY: CITY MODELING FROM SATELLITE IMAGES WITH DEEP IMPLICIT OCCUPANCY FIELDS

C. Stucker, B. Ke, Y. Yue, S. Huang, I. Armeni, K. Schindler
2022 ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
High-resolution optical satellite sensors, combined with dense stereo algorithms, have made it possible to reconstruct 3D city models from space.  ...  especially w.r.t. building reconstruction, featuring intricate roof details, smooth surfaces, and straight, regular outlines.  ...  So far, implicit representations have been explored to model the 3D geometry of local shapes (Genova et al., 2019 , Genova et al., 2020) , single objects (Park et al., 2019, Atzmon and Lipman, 2020)  ... 
doi:10.5194/isprs-annals-v-2-2022-193-2022 fatcat:prbro2ztdvbi3lnmmshvwrk3q4

Piecewise-planar reconstruction using two views

Michel Antunes, João P. Barreto, Urbano Nunes
2016 Image and Vision Computing  
The 3D space is sampled by a set of virtual cut planes that intersect the baseline of the stereo rig and implicitly define possible pixel correspondences across views.  ...  The PEARL algorithm alternates between a discrete optimization step, which merges planar surface hypotheses and discards detections with poor support, and a continuous optimization step, which refines  ...  This work was also supported by FCT and the COMPETE program (co-funded by FEDER) under the project grant AMS-HMI12:RECI/EEI-AUT/0181/2012.  ... 
doi:10.1016/j.imavis.2015.11.008 fatcat:etkimm445fcezl6fqhul3flx2u

ImpliCity: City Modeling from Satellite Images with Deep Implicit Occupancy Fields [article]

Corinne Stucker, Bingxin Ke, Yuanwen Yue, Shengyu Huang, Iro Armeni, Konrad Schindler
2022 arXiv   pre-print
High-resolution optical satellite sensors, combined with dense stereo algorithms, have made it possible to reconstruct 3D city models from space.  ...  w.r.t. building reconstruction, featuring intricate roof details, smooth surfaces, and straight, regular outlines.  ...  So far, implicit representations have been explored to model the 3D geometry of local shapes (Genova et al., 2019 , Genova et al., 2020) , single objects (Park et al., 2019, Atzmon and Lipman, 2020)  ... 
arXiv:2201.09968v3 fatcat:ycv54iu665gnjpacmcy76g6q2m

Author Index

2010 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition  
3D Deformable Surface Reconstruction Torsello, Andrea A Game-Theoretic Approach to Fine Surface Registration without Initial Motion Estimation Object Detection via Boundary Structure Segmentation Detecting  ...  Building Reconstruction using Manhattan-World Grammars Varnavas, Andreas Workshop: Dense Photometric Stereo Reconstruction on Many Core GPUs Varol, Aydin Simultaneous Point Matching and 3D Deformable Surface  ... 
doi:10.1109/cvpr.2010.5539913 fatcat:y6m5knstrzfyfin6jzusc42p54

PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitization [article]

Shunsuke Saito, Zeng Huang, Ryota Natsume, Shigeo Morishima, Angjoo Kanazawa, Hao Li
2019 arXiv   pre-print
We introduce Pixel-aligned Implicit Function (PIFu), a highly effective implicit representation that locally aligns pixels of 2D images with the global context of their corresponding 3D object.  ...  Using PIFu, we propose an end-to-end deep learning method for digitizing highly detailed clothed humans that can infer both 3D surface and texture from a single image, and optionally, multiple input images  ...  ability to digitize and understand 3D objects in the wild.  ... 
arXiv:1905.05172v3 fatcat:aq7mo4wt6nea5cpka24azjswba

A Gaussian measurement model for local interest point based 6 DOF pose estimation

Thilo Grundmann, Wendelin Feiten, Georg v. Wichert
2011 2011 IEEE International Conference on Robotics and Automation  
In this paper we shortly describe a model based object recognition and localization system.  ...  We construct a Gaussian approximation of the resulting pose error using the implicit function theorem. It is then used as a proposal density for importance sampling.  ...  The KIT object modeling center IOMOS [19] is used to acquire stereo images and a precise 3D surface point cloud for each of the objects to be modeled.  ... 
doi:10.1109/icra.2011.5980284 dblp:conf/icra/GrundmannFW11 fatcat:xoxb6uwt4neyhl3d5z2deikc7y

Extracting 3D Scene-Consistent Object Proposals and Depth from Stereo Images [chapter]

Michael Bleyer, Christoph Rhemann, Carsten Rother
2012 Lecture Notes in Computer Science  
occupancy of 3D space and gravity of objects.  ...  Our main contribution is to introduce the concept of 3D scene-consistency into stereo matching. We show that this concept is beneficial for both tasks, object extraction and depth estimation.  ...  Given a stereo pair (a), our algorithm jointly estimates a 3D reconstruction (b) and object maps (c,d) using physics-based reasoning.  ... 
doi:10.1007/978-3-642-33715-4_34 fatcat:i2bagzsf5vbuvhfpmeaqutlhz4

Technical Report: Co-learning of geometry and semantics for online 3D mapping [article]

Marcela Carvalho, Maxime Ferrera, Alexandre Boulch, Julien Moras, Bertrand Le Saux, Pauline Trouvé-Peloux
2019 arXiv   pre-print
Its inputs are an image and a raw depth map produced from a pair of images by standard stereo vision.  ...  In this paper, we address 3D semantic reconstruction for autonomous navigation using co-learning of depth map and semantic segmentation.  ...  Hence, the zero crossing is an implicit representation of the surfaces of the objects present in the scene and a Marching Cube algorithm is used to recovers the mesh.  ... 
arXiv:1911.01082v1 fatcat:ynsc6vrnwnfs5lnhwxe3az62iy
« Previous Showing results 1 — 15 out of 1,389 results