Filters








61,427 Hits in 5.0 sec

A Segmentation Based Variational Model for Accurate Optical Flow Estimation [chapter]

Li Xu, Jianing Chen, Jiaya Jia
2008 Lecture Notes in Computer Science  
First, we partition the input images and integrate the segmentation information into a variational model where each of the segments is constrained by an affine motion.  ...  Extensive experiments show that the proposed method not only produces quantitatively accurate optical flow estimates but also preserves sharp motion boundaries, which makes the optical flow result usable  ...  Acknowledgements We thank Guofeng Zhang for his comments on this paper. This work was fully supported by a grant from the Research Grants Council of Hong Kong (Project No. 412708).  ... 
doi:10.1007/978-3-540-88682-2_51 fatcat:3crwupvplzez5bzfpdq6n4ek7u

Video Deblurring via Semantic Segmentation and Pixel-Wise Non-Linear Kernel [article]

Wenqi Ren, Jinshan Pan, Xiaochun Cao, Ming-Hsuan Yang
2017 arXiv   pre-print
In this paper, we exploit semantic segmentation in each blurry frame to understand the scene contents and use different motion models for image regions to guide optical flow estimation.  ...  Optical flow can be used for kernel estimation since it predicts motion trajectories.  ...  Second, we exploit semantic segmentation to account for occlusions and blurry edges for accurate optical flow estimation.  ... 
arXiv:1708.03423v1 fatcat:ztfjqbiyajcubatwzihhvmbrxe

Video Deblurring via Semantic Segmentation and Pixel-Wise Non-linear Kernel

Wenqi Ren, Jinshan Pan, Xiaochun Cao, Ming-Hsuan Yang
2017 2017 IEEE International Conference on Computer Vision (ICCV)  
In this paper, we exploit semantic segmentation in each blurry frame to understand the scene contents and use different motion models for image regions to guide optical flow estimation.  ...  Optical flow can be used for kernel estimation since it predicts motion trajectories.  ...  Second, we exploit semantic segmentation to account for occlusions and blurry edges for accurate optical flow estimation.  ... 
doi:10.1109/iccv.2017.123 dblp:conf/iccv/RenPC017 fatcat:cvz7fjcrwrfqzipu63tzxre5vm

Object Segmentation Tracking from Generic Video Cues [article]

Amirhossein Kardoost, Sabine Müller, Joachim Weickert, Margret Keuper
2020 arXiv   pre-print
We propose a light-weight variational framework for online tracking of object segmentations in videos based on optical flow and image boundaries.  ...  While high-end computer vision methods on this task rely on sequence specific training of dedicated CNN architectures, we show the potential of a variational model, based on generic video information from  ...  All further results are based on this setting. 2) Flow Estimation Methods: Optical flow information is a central component of our model.  ... 
arXiv:1910.02258v3 fatcat:aw6wouvesbfgjgg2vihfqcnkja

Generalized Video Deblurring for Dynamic Scenes [article]

Tae Hyun Kim, Kyoung Mu Lee
2015 arXiv   pre-print
Therefore, we propose a single energy model that simultaneously estimates optical flows and latent frames to solve our deblurring problem.  ...  By minimizing the proposed energy function, we achieve significant improvements in removing blurs and estimating accurate optical flows in blurry frames.  ...  By estimating a pixel-wise kernel using optical flows, we handled general blurs. Thus, we proposed a new energy model that estimates optical flows and latent frames, jointly.  ... 
arXiv:1507.02438v1 fatcat:rg3oicetsbgfhils76n5kioxx4

Generalized video deblurring for dynamic scenes

Tae Hyun Kim, Kyoung Mu Lee
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
Therefore, we propose a single energy model that simultaneously estimates optical flows and latent frames to solve our deblurring problem.  ...  By minimizing the proposed energy function, we achieve significant improvements in removing blurs and estimating accurate optical flows in blurry frames.  ...  By estimating a pixel-wise kernel using optical flows, we handled general blurs. Thus, we proposed a new energy model that estimates optical flows and latent frames, jointly.  ... 
doi:10.1109/cvpr.2015.7299181 dblp:conf/cvpr/KimL15a fatcat:itv3caqg7rbctclrfweogvkxcy

SS-SF: Piecewise 3D Scene Flow Estimation with Semantic Segmentation

Cheng Feng, Long Ma, Congxuan Zhang, Zhen Chen, Liyue Ge, Shaofeng Jiang
2021 IEEE Access  
Second, we plan a novel energy function to optimize the initial mappings by using a semantic segmentation constraint term to regularize the classical scene flow model, which the optimized mappings are  ...  In order to address the issue of edge-blurring and improve the accuracy and robustness of scene flow estimation under motion occlusions, we in this article propose a piecewise 3D scene flow estimation  ...  For example, in order to gain a robust computation scheme, Gottfried et al. [39] investigated a variational framework for RGB-D scene flow estimation.  ... 
doi:10.1109/access.2021.3055939 fatcat:nhefqnwjmzh2vd3ncbbaajwzwa

Dense, accurate optical flow estimation with piecewise parametric model

Jiaolong Yang, Hongdong Li
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
This paper proposes a simple method for estimating dense and accurate optical flow field. It revitalizes an early idea of piecewise parametric flow model.  ...  A key innovation is that, we fit a flow field piecewise to a variety of parametric models, where the domain of each piece (i.e., each piece's shape, position and size) is determined adaptively, while at  ...  JY is a CSC-cofunded BIT-ANU joint PhD student. The authors would like to thank AC and anonymous reviewers for invaluable comments.  ... 
doi:10.1109/cvpr.2015.7298704 dblp:conf/cvpr/YangL15 fatcat:dqjskno26bhtjfb5vnszqnqhha

Variational Motion Segmentation with Level Sets [chapter]

Thomas Brox, Andrés Bruhn, Joachim Weickert
2006 Lecture Notes in Computer Science  
We suggest a variational method for the joint estimation of optic flow and the segmentation of the image into regions of similar motion.  ...  It need not fear a quantitative comparison to pure optic flow estimation techniques: For the popular Yosemite sequence with clouds we obtain the currently most accurate result.  ...  Acknowledgements We thank Luis Garrido (University Pompeu Fabral, Barcelona, Spain) for raising the question if there is an error in the ground truth of the Yosemite sequence.  ... 
doi:10.1007/11744023_37 fatcat:dw2cs3gcj5c7tlsfmvtuoaioam

A General Dense Image Matching Framework Combining Direct and Feature-Based Costs

Jim Braux-Zin, Romain Dupont, Adrien Bartoli
2013 2013 IEEE International Conference on Computer Vision  
Dense motion field estimation (typically optical flow, stereo disparity and surface registration) is a key computer vision problem.  ...  The feature-based cost is built around a novel robust distance function that handles keypoints and "weak" features such as segments.  ...  Their method remains a reference in optical flow estimation for its robustness and accuracy, but we identified several limitations.  ... 
doi:10.1109/iccv.2013.30 dblp:conf/iccv/Braux-ZinDB13 fatcat:fmis34zujvg43bpmie4vt42vga

A Feature-Based Approach for Determining Dense Long Range Correspondences [chapter]

Josh Wills, Serge Belongie
2004 Lecture Notes in Computer Science  
For this reason, to achieve dense optical flow for image sequences with large inter-frame disparity, we propose a two stage process in which a planar model is used to get an approximation for the segmentation  ...  Planar motion models can provide gross motion estimation and good segmentation for image pairs with large inter-frame disparity.  ...  We would like to thank Sameer Agarwal, Charless Fowlkes and Ben Ochoa for helpful discussions. The images in Figures  ... 
doi:10.1007/978-3-540-24672-5_14 fatcat:crkdiz55afhvvnjg7jfjmzxl5y

Motion Estimation Using Region-Level Segmentation and Extended Kalman Filter for Autonomous Driving

Hongjian Wei, Yingping Huang, Fuzhi Hu, Baigan Zhao, Zhiyang Guo, Rui Zhang
2021 Remote Sensing  
We first use a region-level segmentation to accurately locate the object region for the latter two stages.  ...  In the stage of parameter estimate, we develop a relative motion model of the ego-vehicle and the object, and accordingly establish an EKF model for point tracking and parameter estimate.  ...  In this study, we leverage a region-level segmentation to accurately locate object regions for tracking and parameter estimate.  ... 
doi:10.3390/rs13091828 fatcat:pbxdishdtnh4tdo5uhxydw7rhi

Efficient Uncertainty Estimation for Semantic Segmentation in Videos [article]

Po-Yu Huang, Wan-Ting Hsu, Chun-Yueh Chiu, Ting-Fan Wu, Min Sun
2018 arXiv   pre-print
A deep learning model can't be applied in real applications if we don't know whether the model is certain about the decision or not.  ...  For real-time applications such as a self-driving car system, which needs to obtain the prediction and the uncertainty as fast as possible, so that MC dropout becomes impractical.  ...  For some regions that contain fast moving objects or occlusion, the optical flow may not be accurate.  ... 
arXiv:1807.11037v1 fatcat:trbnkqwcevdsnara2i2dszsyre

Dynamic Texture Detection Based on Motion Analysis

Sándor Fazekas, Tomer Amiaz, Dmitry Chetverikov, Nahum Kiryati
2008 International Journal of Computer Vision  
Motion estimation is usually based on the brightness constancy assumption.  ...  Accurate segmentation into regions of static and dynamic texture is achieved using a level set scheme.  ...  Acknowledgements This research was supported partly by MUS-CLE: Multimedia Understanding through Semantics, Computation and Learning, a European Network of Excellence funded by the EC 6th Framework IST  ... 
doi:10.1007/s11263-008-0184-y fatcat:ankxf2eazfentlnmkxo26jkpfy

Real-time object tracking and segmentation using adaptive color snake model

Kap-Ho Seo, Ju-Jang Lee
2005 31st Annual Conference of IEEE Industrial Electronics Society, 2005. IECON 2005.  
An active contour model, snake, was developed as a useful segmenting and tracking tool for rigid or non-rigid objects.  ...  In this paper, the development of new snake model called "adaptive color snake model (ACSM)" for segmentation and tracking is introduced. The simple operation makes the algorithm runs in real-time.  ...  As the camera moves, the system generates many variations between successive images. Optical flow-based approach is not suitable for these situations.  ... 
doi:10.1109/iecon.2005.1569195 fatcat:dg2c2jyvknft5c34m5kzufpcfm
« Previous Showing results 1 — 15 out of 61,427 results