Filters








84 Hits in 7.3 sec

Nonrigid Optical Flow Ground Truth for Real-World Scenes With Time-Varying Shading Effects

Wenbin Li, Darren Cosker, Zhihan Lv, Matthew Brown
2017 IEEE Robotics and Automation Letters  
In this paper we present a dense ground truth dataset of nonrigidly deforming real-world scenes.  ...  Our capture protocol enables us to acquire real-world deforming objects with realistic photometric effects - such as blur and illumination change - as well as occlusion and complex deformations.  ...  Nonrigid Optical Flow Ground Truth for Real-World Scenes with Time-Varying Shading Effects Wenbin Li, Darren Cosker, Zhihan Lv and Matthew Brown Abstract-In this paper we present a dense ground truth dataset  ... 
doi:10.1109/lra.2016.2592513 dblp:journals/ral/0002CLB17 fatcat:njszkohgzja2rcvgpivwgkayeu

What Makes Good Synthetic Training Data for Learning Disparity and Optical Flow Estimation?

Nikolaus Mayer, Eddy Ilg, Philipp Fischer, Caner Hazirbas, Daniel Cremers, Alexey Dosovitskiy, Thomas Brox
2018 International Journal of Computer Vision  
Yet, for many computer vision problems, such as stereo or optical flow estimation, this approach is not feasible because humans cannot manually enter a pixel-accurate flow field.  ...  As a result, the research challenge shifts from devising algorithms to creating suitable and abundant training data for supervised learning. How to efficiently create such training data?  ...  : a 3D dataset rendered from true 3D models, with ground truth for stereo disparity, optical flow, and for the full scene flow task (Mayer et al, 2016) .  ... 
doi:10.1007/s11263-018-1082-6 fatcat:pko3jsec5bcqhgqlxwa23by2fi

HDM-Net: Monocular Non-Rigid 3D Reconstruction with Learned Deformation Model [article]

Vladislav Golyanik and Soshi Shimada and Kiran Varanasi and Didier Stricker
2019 arXiv   pre-print
HDM-Net learns different reconstruction cues including texture-dependent surface deformations, shading and contours.  ...  In this work, we propose a new hybrid approach for monocular non-rigid reconstruction which we call Hybrid Deformation Model Network (HDM-Net).  ...  However, this estimate does not include often prohibitive computation time of dense correspondences with multi-frame optical flow methods such as [64] .  ... 
arXiv:1803.10193v2 fatcat:szrra6fw6vat5m7y7gjulzoiya

3D Trajectory Reconstruction under Perspective Projection

Hyun Soo Park, Takaaki Shiratori, Iain Matthews, Yaser Sheikh
2015 International Journal of Computer Vision  
We note that such basis vectors are also coordinate independent, which allows us to directly use camera poses estimated from stationary areas in the scene (in contrast to nonrigid structure from motion  ...  each image and the relative camera poses at each time instant.  ...  We identify real-world scenarios where high reconstruction accuracy is guaranteed via the geometric analysis and apply our algorithm to reconstruct the time-varying 3D structure of the scenes.  ... 
doi:10.1007/s11263-015-0804-2 fatcat:cl3zqo33nbgexohkcchc5yncde

Better Together: Joint Reasoning for Non-rigid 3D Reconstruction with Specularities and Shading [article]

Qi Liu-Yin, Rui Yu, Lourdes Agapito, Andrew Fitzgibbon, Chris Russell
2017 arXiv   pre-print
We demonstrate the use of shape-from-shading (SfS) to improve both the quality and the robustness of 3D reconstruction of dynamic objects captured by a single camera.  ...  Moving beyond Lambertian S f S , we propose a general approach that models both specularities and shading while simultaneously tracking and reconstructing general dynamic objects.  ...  Although, we focused upon the challenging problem of reconstruction from a single RGB camera, we have also shown that joint reasoning about the motions of deformable objects and shape-from-shading can  ... 
arXiv:1708.01654v1 fatcat:gp2wz3senncmbibikcidiijesu

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  
For the past few years, researchers have spent significant effort to develop algorithms to capture 3D models with RGB-D cameras.  ...  for existing systems.  ...  Quantitative evaluation is suitable for comparisons on synthetic scenes (e.g., the ICL-NUIM dataset), but is challenging on real-world scenes as there is typically no ground-truth surface model.  ... 
doi:10.1007/s41095-021-0250-8 fatcat:z6ywcn4zujbptjaqrwkwhoyquu

Effects of surface reflectance and 3D shape on perceived rotation axis

K. Doerschner, O. Yilmaz, G. Kucukoglu, R. W. Fleming
2013 Journal of Vision  
Surface specularity distorts the optic flow generated by a moving object in a way that provides important cues for identifying surface material properties (Doerschner, Fleming et al., 2011) .  ...  In general, our findings are in line with earlier research that demonstrated that shape from motion can be extracted based on several different types of optical deformation  ...  Acknowledgments This work was supported by a Marie Curie International Reintegration Grant (239494) Because of the appearance distortion in specular flow, the corresponding (across frames) specular  ... 
doi:10.1167/13.11.8 pmid:24023275 fatcat:bmtzj63jfvdgnm3cnh7hr2mp6y

Shape, motion, and optical cues to stiffness of elastic objects

Vivian C. Paulun, Filipp Schmidt, Jan Jaap R. van Assen, Roland W. Fleming
2017 Journal of Vision  
Softness ratings for static, unperturbed cubes varied substantially and systematically depending on the optical properties.  ...  However, when animated, the ratings were again dominated by the extent of the deformation, and the effect of optical appearance was negligible.  ...  Acknowledgments We would like to thank Eugen Prokott for assistance in designing the materials and Maarten Wijntjes for helpful comments on a previous version of this manuscript.  ... 
doi:10.1167/17.1.20 pmid:28114494 fatcat:2aemjcmgmrahboeoqjuesujyzi

FAUST: Dataset and Evaluation for 3D Mesh Registration

Federica Bogo, Javier Romero, Matthew Loper, Michael J. Black
2014 2014 IEEE Conference on Computer Vision and Pattern Recognition  
While the challenges of real-world data registration are not present in existing synthetic datasets, establishing ground-truth correspondences for real 3D scans is difficult.  ...  We find that current shape registration methods have trouble with this real-world data. The dataset and evaluation website are available for research purposes at http://faust.is.tue.mpg.de.  ...  Holderness for her help with data acquisition.  ... 
doi:10.1109/cvpr.2014.491 dblp:conf/cvpr/Bogo0LB14 fatcat:wwswabfllnahhodw32imwwq2ue

Three-Dimensional Tissue Deformation Recovery and Tracking

Peter Mountney, Danail Stoyanov, Guang-Zhong Yang
2010 IEEE Signal Processing Magazine  
(e) Schematic of a laparoscope with imaging optics observing a sample of tissue in 3D.  ...  These optically based techniques provide a unique opportunity for recovering surface deformation of the softtissue without the need of additional instrumentation.  ...  They are not suitable for real-time applications as the deformation is dealt with in an offline, global optimization step. Nonrigid Structure from Motion has been applied to the heart [67] .  ... 
doi:10.1109/msp.2010.936728 fatcat:2zzk2rfsbbc6jck6j3jkkztzo4

Image-based Tracking of Deformable Surfaces [chapter]

Anna Hilsmann, David C., Peter Eisert
2011 Object Tracking  
For deformable surfaces, the intensity of a scene point varies when the surface normals change.  ...  Usually, two successive frames in a natural video sequence do not only differ geometrically, but also the intensity of a scene point can vary due to changes in the scene lighting, shading properties etc  ...  This chapter is distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike-3.0 License, which permits use, distribution and reproduction for non-commercial purposes, provided  ... 
doi:10.5772/14344 fatcat:govg5o4dpbdujn5xiolru3nh3y

The perceptual dimensions of natural dynamic flow

Yaniv Morgenstern, Daniel J. Kersten
2017 Journal of Vision  
Acknowledgments We thank Kendrick Kay for helpful discussions on data analysis, and Shinho Cho, Erik Wingerson, Hanlin Zhu, and the reviewers for helpful comments.  ...  The mean ratings relative to ground truth for stimuli from the minor-and major-contextual-effects subsets.  ...  The mean ratings relative to ground truth for stimuli from the minor-and major-contextual-effects subsets.  ... 
doi:10.1167/17.12.7 pmid:29049594 fatcat:yueyimirtrahzkkmyno7mgvkoi

Roto++

Wenbin Li, Fabio Viola, Jonathan Starck, Gabriel J. Brostow, Neill D. F. Campbell
2016 ACM Transactions on Graphics  
Finally, we also provide a new, professionally rotoscoped dataset that enables truly representative, real-world evaluation of rotoscoping methods.  ...  In this work, we study the particular task of professional rotoscoping for high-end, live action movies and propose a new framework that works with roto-artists to accelerate the workflow and improve their  ...  All data-sets were prepared by Courtney Pryce with special thanks to John Devasahayam and the PixStone Images team for turning around a vast amount of work in such a short period of time.  ... 
doi:10.1145/2897824.2925973 fatcat:tgt6twwfnnfw3bjws5wu2jop2m

Unwrap mosaics

Alex Rav-Acha, Pushmeet Kohli, Carsten Rother, Andrew Fitzgibbon
2008 ACM SIGGRAPH 2008 papers on - SIGGRAPH '08  
attaching effects layers to deforming objects.  ...  Abstract We introduce a new representation for video which facilitates a number of common editing tasks.  ...  Acknowledgements Discussions with many people have contributed to this work in its long development.  ... 
doi:10.1145/1399504.1360616 fatcat:3rvqrgb4r5fplbelf2inxhfjvu

Unwrap mosaics

Alex Rav-Acha, Pushmeet Kohli, Carsten Rother, Andrew Fitzgibbon
2008 ACM Transactions on Graphics  
attaching effects layers to deforming objects.  ...  Abstract We introduce a new representation for video which facilitates a number of common editing tasks.  ...  Acknowledgements Discussions with many people have contributed to this work in its long development.  ... 
doi:10.1145/1360612.1360616 fatcat:yazep25nenduxnppzqmfm5i5z4
« Previous Showing results 1 — 15 out of 84 results