Filters








18,397 Hits in 5.2 sec

Dynamic Body VSLAM with Semantic Constraints [article]

N. Dinesh Reddy, Prateek Singhal, Visesh Chari, K. Madhava Krishna
2015 arXiv   pre-print
optimization process that does iterative refinement of 3D structure and camera / object motion trajectories.  ...  We show results on the challenging KITTI urban dataset for accuracy of motion segmentation and reconstruction of the trajectory and shape of moving objects relative to ground truth.  ...  TRAJECTORY ESTIMATION The motion segmented images of the static world and moving objects are then used as input to localize and map each object independently.  ... 
arXiv:1504.07269v1 fatcat:nvgjgwacznecvjvkkveljrvuna

Dynamic body VSLAM with semantic constraints

N. Dinesh Reddy, Prateek Singhal, Visesh Chari, K. Madhava Krishna
2015 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)  
optimization process that does iterative refinement of 3D structure and camera / object motion trajectories.  ...  We show results on the challenging KITTI urban dataset for accuracy of motion segmentation and reconstruction of the trajectory and shape of moving objects relative to ground truth.  ...  Right 3d reconstruction with overlaid semantic map and trajectories of the moving objects and camera.  ... 
doi:10.1109/iros.2015.7353626 dblp:conf/iros/ReddySCK15 fatcat:3xacj3eptnajbjdfqsdig374vy

Realtime motion segmentation based multibody visual SLAM

Abhijit Kundu, K. Madhava Krishna, C. V. Jawahar
2010 Proceedings of the Seventh Indian Conference on Computer Vision, Graphics and Image Processing - ICVGIP '10  
The proposed framework of multibody visual SLAM allows choosing between full 3D reconstruction or simply tracking of the moving objects, which adds flexibility to the system, for scenes containing non-rigid  ...  We adopt a multibody Structure from Motion (SfM) approach, which is the generalization of classical SfM to dynamic scenes with multiple rigidly moving objects.  ...  The integrated system allows simultaneous online 3D reconstruction and tracking of multiple moving objects using only a single monocular camera.  ... 
doi:10.1145/1924559.1924593 dblp:conf/icvgip/KunduKJ10 fatcat:tzaqb3bm7bf6lcc6i7qx2zqlei

Realtime multibody visual SLAM with a smoothly moving monocular camera

Abhijit Kundu, K Madhava Krishna, C. V. Jawahar
2011 2011 International Conference on Computer Vision  
Tracking and reconstruction is preceded by motion segmentation and detection which makes use of efficient geometric constraints to avoid difficult degenerate motions, where objects move in the epipolar  ...  This enables building of a unified dynamic 3D map of scenes involving multiple moving objects.  ...  It shows the camera trajectory and 3D structure of static background. Reconstruction and the 3D trajectory of a moving car in the scene as produced by the system are also shown.  ... 
doi:10.1109/iccv.2011.6126482 dblp:conf/iccv/KunduKJ11 fatcat:57v6r7h645f57cleuoqpyrk5la

3D Trajectory Reconstruction of Dynamic Objects Using Planarity Constraints [article]

Sebastian Bullinger, Christoph Bodensteiner, Michael Arens, Rainer Stiefelhagen
2017 arXiv   pre-print
We present a method to reconstruct the three-dimensional trajectory of a moving instance of a known object category in monocular video data.  ...  Due to the lack of suitable benchmark datasets we present a new dataset to evaluate the quality of reconstructed three-dimensional object trajectories.  ...  [16] exploit the non-accidentalness principle in the context of independently moving objects. Yuan et al.  ... 
arXiv:1711.06136v1 fatcat:fhldbk2jk5eu3gvqiuoqvhebx4

Trajectory Space: A Dual Representation for Nonrigid Structure from Motion

I Akhter, Y Sheikh, S Khan, T Kanade
2011 IEEE Transactions on Pattern Analysis and Machine Intelligence  
The principal advantage of expressing deforming 3D structure in trajectory space is that we can define an object independent basis.  ...  We further show that the temporal smoothness in 3D trajectories alone can be used for recovering nonrigid structure from a moving camera.  ...  ACKNOWLEDGMENTS This research was partly supported by a grant from the Higher Education Commission of Pakistan. The authors acknowledge Fernando De La Torre for useful discussions.  ... 
doi:10.1109/tpami.2010.201 pmid:21079275 fatcat:gbhjadyblfc7nlhxpgx5zntmza

Incomplete 3D motion trajectory segmentation and 2D-to-3D label transfer for dynamic scene analysis

Cansen Jiang, Danda Pani Paudel, Yohan Fougerolle, David Fofi, Cedric Demonceaux
2017 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)  
We introduce a complete framework for static-map and dynamic objects' reconstruction, as well as semantic scene understanding for a calibrated and moving 2D-3D camera setup.  ...  The key information for these tasks lies on semantic labels of the scene parts and the motion trajectories of the dynamic objects.  ...  Therefore, this work offers a general scene understanding and modelling framework of dynamic environments using a moving 2D-3D camera platform.  ... 
doi:10.1109/iros.2017.8202214 dblp:conf/iros/JiangPFFD17 fatcat:5t67srz4fzf5lgluu7ec755jbm

Nonrigid Structure from Motion in Trajectory Space

Ijaz Akhter, Yaser Sheikh, Sohaib Khan, Takeo Kanade
2008 Neural Information Processing Systems  
We show that generic bases over trajectories, such as the Discrete Cosine Transform (DCT) basis, can be used to compactly describe most real motions.  ...  In contrast, we propose that the evolving 3D structure be described by a linear combination of basis trajectories.  ...  Acknowledgements This research was partially supported by a grant from the Higher Education Commission of Pakistan. The authors would like to acknowledge Fernando De La Torre for useful discussions.  ... 
dblp:conf/nips/AkhterSKK08 fatcat:ippmpmn7cvcopdabozbwrqrlui

3D Trajectory Reconstruction under Perspective Projection

Hyun Soo Park, Takaaki Shiratori, Iain Matthews, Yaser Sheikh
2015 International Journal of Computer Vision  
We represent a 3D trajectory using a linear combination of compact trajectory basis vectors, such as the discrete cosine transform basis, that have been shown to approximate object independence.  ...  We present an algorithm to reconstruct the 3D trajectory of a moving point from its correspondence in a collection of temporally non-coincidental 2D perspective images, given the time of capture that produced  ...  . (2008) identified that the trajectory basis vectors are object independent.  ... 
doi:10.1007/s11263-015-0804-2 fatcat:cl3zqo33nbgexohkcchc5yncde

3D Vehicle Trajectory Reconstruction in Monocular Video Data Using Environment Structure Constraints [chapter]

Sebastian Bullinger, Christoph Bodensteiner, Michael Arens, Rainer Stiefelhagen
2018 Lecture Notes in Computer Science  
Our algorithm achieves an average reconstruction-to-groundtruth-trajectory distance of 0.31 meter using this dataset.  ...  We present a framework to reconstruct three-dimensional vehicle trajectories using monocular video data.  ...  [18] exploit the nonaccidentalness principle in the context of independently moving objects. Yuan et al.  ... 
doi:10.1007/978-3-030-01249-6_3 fatcat:qpghvzkx25ee7mm6zoieznut5a

3D reconstruction of a smooth articulated trajectory from a monocular image sequence

Hyun Soo Park, Yaser Sheikh
2011 2011 International Conference on Computer Vision  
We simultaneously apply activity-independent spatial and temporal constraints, in the form of fixed 3D distance to the parent trajectory and smooth 3D motion.  ...  A geometric analysis of the reconstruction of articulated trajectories is also presented and a measure of the reconstructibility of an articulated trajectory is proposed.  ...  Head and torso are used as the root for playing card motion and yoga motion, respectively. Related Work Points on objects involve spatial and temporal constraints in general. Bregler et al.  ... 
doi:10.1109/iccv.2011.6126243 dblp:conf/iccv/ParkS11 fatcat:whp2mlyj5be55ndclxd6gofgsq

Trajectory triangulation: 3D reconstruction of moving points from a monocular image sequence

S. Avidan, A. Shashua
2000 IEEE Transactions on Pattern Analysis and Machine Intelligence  
AbstractÐWe consider the problem of reconstructing the 3D coordinates of a moving point seen from a monocular moving camera, i.e., to reconstruct moving objects from line-of-sight measurements only.  ...  Thus static scenes become a particular case of a more general task of reconstructing scenes rich with moving objects (where an object could be a single point).  ...  However, consider the next (natural) question in this context: Can one reconstruct the 3D coordinates of a (single) point on a moving object?  ... 
doi:10.1109/34.845377 fatcat:2mjhzzciujbj3goe4y2r6tb7pa

A Low-Rank Constraint for Parallel Stereo Cameras [chapter]

Christian Cordes, Hanno Ackermann, Bodo Rosenhahn
2013 Lecture Notes in Computer Science  
Stereo-camera systems enjoy wide popularity since they provide more restrictive constraints for 3d-reconstruction.  ...  Given an image sequence taken by parallel stereo cameras, a low-rank constraint is derived on the measurement data.  ...  Given only two, three or four images taken by cameras in general configuration, 3D-reconstructions can be computed using the epipolar constraint.  ... 
doi:10.1007/978-3-642-40602-7_4 fatcat:oadntqzlnbckri7icfimqngacu

Stereo 3D Object Trajectory Reconstruction [article]

Sebastian Bullinger, Christoph Bodensteiner, Michael Arens, Rainer Stiefelhagen
2018 arXiv   pre-print
We present a method to reconstruct the three-dimensional trajectory of a moving instance of a known object category using stereo video data.  ...  As opposed to monocular trajectory reconstruction approaches, our method shows no degenerated cases. We evaluate our approach using publicly available video data of vehicles in urban scenes.  ...  Without loss of generality, we describe motion trajectory reconstructions of single objects. We apply SfM [18, 22] to object and background images as shown in Fig. 1 .  ... 
arXiv:1808.09297v1 fatcat:t7yajb744nf23ig3bieipgwnr4

TrackCam

Shuaicheng Liu, Jue Wang, Sunghyun Cho, Ping Tan
2014 ACM Transactions on Graphics  
In this work we propose a system to generate realistic, 3D-aware tracking shots from consumer videos.  ...  We also introduce a pseudo 3D approach for relative depth estimation to avoid expensive 3D reconstruction for improved robustness and a wider application range.  ...  Ozden et al. [2004] addressed the problem of reconstructing 3D trajectories of moving objects from a monocular video using generic constraints for the analysis of dynamic scenes, with an emphasis on determining  ... 
doi:10.1145/2661229.2661272 fatcat:a33pcmncvbbt7ijh7yvp4bmgma
« Previous Showing results 1 — 15 out of 18,397 results