�ݺ�ߣshows by User: BrunoConejobconejogm

�ݺ�ߣshows by User: BrunoConejobconejogm / http://www.slideshare.net/images/logo.gif �ݺ�ߣshows by User: BrunoConejobconejogm / Wed, 21 Jan 2015 06:20:15 GMT �ݺ�ߣShare feed for �ݺ�ߣshows by User: BrunoConejobconejogm Inference by Learning: Speeding-?�\up Graphical Model Optimization via a Coarse-?to-?Fine Cascade of Pruning Classifiers (NIPS 2014) /slideshow/inference-by-learning/43737674 nips2014ibylslides-150121062015-conversion-gate02
We propose a general and versatile framework that significantly speeds-up graphical model optimization while maintaining an excellent solution accuracy. The proposed approach, refereed as Inference by Learning or in short as IbyL, relies on a multi-scale pruning scheme that progressively reduces the solution space by use of a coarse-to-fine cascade of learnt classifiers. We thoroughly experiment with classic computer vision related MRF problems, where our novel framework constantly yields a significant time speed-up (with respect to the most efficient inference methods) and obtains a more accurate solution than directly optimizing the MRF. We make our code available on-line: http://imagine.enpc.fr/~conejob/ibyl/index.html]]>
We propose a general and versatile framework that significantly speeds-up graphical model optimization while maintaining an excellent solution accuracy. The proposed approach, refereed as Inference by Learning or in short as IbyL, relies on a multi-scale pruning scheme that progressively reduces the solution space by use of a coarse-to-fine cascade of learnt classifiers. We thoroughly experiment with classic computer vision related MRF problems, where our novel framework constantly yields a significant time speed-up (with respect to the most efficient inference methods) and obtains a more accurate solution than directly optimizing the MRF. We make our code available on-line: http://imagine.enpc.fr/~conejob/ibyl/index.html]]> Wed, 21 Jan 2015 06:20:15 GMT /slideshow/inference-by-learning/43737674 BrunoConejobconejogm@slideshare.net(BrunoConejobconejogm) Inference by Learning: Speeding-?�\up Graphical Model Optimization via a Coarse-?to-?Fine Cascade of Pruning Classifiers (NIPS 2014) BrunoConejobconejogm We propose a general and versatile framework that significantly speeds-up graphical model optimization while maintaining an excellent solution accuracy. The proposed approach, refereed as Inference by Learning or in short as IbyL, relies on a multi-scale pruning scheme that progressively reduces the solution space by use of a coarse-to-fine cascade of learnt classifiers. We thoroughly experiment with classic computer vision related MRF problems, where our novel framework constantly yields a significant time speed-up (with respect to the most efficient inference methods) and obtains a more accurate solution than directly optimizing the MRF. We make our code available on-line: http://imagine.enpc.fr/~conejob/ibyl/index.html <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/nips2014ibylslides-150121062015-conversion-gate02-thumbnail.jpg?width=120&height=120&fit=bounds" /><br> We propose a general and versatile framework that significantly speeds-up graphical model optimization while maintaining an excellent solution accuracy. The proposed approach, refereed as Inference by Learning or in short as IbyL, relies on a multi-scale pruning scheme that progressively reduces the solution space by use of a coarse-to-fine cascade of learnt classifiers. We thoroughly experiment with classic computer vision related MRF problems, where our novel framework constantly yields a significant time speed-up (with respect to the most efficient inference methods) and obtains a more accurate solution than directly optimizing the MRF. We make our code available on-line: http://imagine.enpc.fr/~conejob/ibyl/index.html

Inference by Learning: Speeding-�ĥ�MӍp Graphical Model Optimization via a Coarse-�ĥ�to-�ĥ�Fine Cascade of Pruning Classifiers (NIPS 2014) from Bruno Conejo (bconejo@gmail.com)

]]> 547 1 https://cdn.slidesharecdn.com/ss_thumbnails/nips2014ibylslides-150121062015-conversion-gate02-thumbnail.jpg?width=120&height=120&fit=bounds presentation 000000 http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Fast Global Stereo Matching Via Energy Pyramid Minimization /slideshow/fast-global-stereo-matching-via-energy-pyramid-minimization/43577898 pcv2014gmeporal-150116032454-conversion-gate02
We define a global matching framework based on energy pyramid, the Global Matching via Energy Pyramid (GM-EP) algorithm, which estimates the disparity map from a single stereo-pair by solving an energy minimization problem. We efficiently address this minimization by globally optimizing a coarse to fine sequence of sparse Conditional Random Fields (CRF) directly defined on the energy. This global discrete optimization approach guarantees that at each scale we obtain a near optimal solution, and we demonstrate its superiority over state of the art image pyramid approaches through application to real stereo-pairs. We conclude that multiscale approaches should be build on energy pyramids rather than on image pyramids.]]>
We define a global matching framework based on energy pyramid, the Global Matching via Energy Pyramid (GM-EP) algorithm, which estimates the disparity map from a single stereo-pair by solving an energy minimization problem. We efficiently address this minimization by globally optimizing a coarse to fine sequence of sparse Conditional Random Fields (CRF) directly defined on the energy. This global discrete optimization approach guarantees that at each scale we obtain a near optimal solution, and we demonstrate its superiority over state of the art image pyramid approaches through application to real stereo-pairs. We conclude that multiscale approaches should be build on energy pyramids rather than on image pyramids.]]> Fri, 16 Jan 2015 03:24:54 GMT /slideshow/fast-global-stereo-matching-via-energy-pyramid-minimization/43577898 BrunoConejobconejogm@slideshare.net(BrunoConejobconejogm) Fast Global Stereo Matching Via Energy Pyramid Minimization BrunoConejobconejogm We define a global matching framework based on energy pyramid, the Global Matching via Energy Pyramid (GM-EP) algorithm, which estimates the disparity map from a single stereo-pair by solving an energy minimization problem. We efficiently address this minimization by globally optimizing a coarse to fine sequence of sparse Conditional Random Fields (CRF) directly defined on the energy. This global discrete optimization approach guarantees that at each scale we obtain a near optimal solution, and we demonstrate its superiority over state of the art image pyramid approaches through application to real stereo-pairs. We conclude that multiscale approaches should be build on energy pyramids rather than on image pyramids. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/pcv2014gmeporal-150116032454-conversion-gate02-thumbnail.jpg?width=120&height=120&fit=bounds" /><br> We define a global matching framework based on energy pyramid, the Global Matching via Energy Pyramid (GM-EP) algorithm, which estimates the disparity map from a single stereo-pair by solving an energy minimization problem. We efficiently address this minimization by globally optimizing a coarse to fine sequence of sparse Conditional Random Fields (CRF) directly defined on the energy. This global discrete optimization approach guarantees that at each scale we obtain a near optimal solution, and we demonstrate its superiority over state of the art image pyramid approaches through application to real stereo-pairs. We conclude that multiscale approaches should be build on energy pyramids rather than on image pyramids.

Fast Global Stereo Matching Via Energy Pyramid Minimization from Bruno Conejo (bconejo@gmail.com)

]]> 759 3 https://cdn.slidesharecdn.com/ss_thumbnails/pcv2014gmeporal-150116032454-conversion-gate02-thumbnail.jpg?width=120&height=120&fit=bounds presentation 000000 http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 https://cdn.slidesharecdn.com/profile-photo-BrunoConejobconejogm-48x48.jpg?cb=1523541458 I am a result-oriented engineer researcher with excellent analytical and organizational skills. My field of expertise is at the boundary of computer vision and machine learning, where I design, implement and publish in top rank conferences algorithms for stereo-matching and graphical models inference. I have demonstrated the ability to quickly learn, apply knowledge and develop new tools, while timely completing tasks in challenging environments. I utilize leadership, communication and interpersonal skills to build successful work relationship. http://www.safran-group.com/spip.php?lang=en https://cdn.slidesharecdn.com/ss_thumbnails/nips2014ibylslides-150121062015-conversion-gate02-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/inference-by-learning/43737674 Inference by Learning:... https://cdn.slidesharecdn.com/ss_thumbnails/pcv2014gmeporal-150116032454-conversion-gate02-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/fast-global-stereo-matching-via-energy-pyramid-minimization/43577898 Fast Global Stereo Mat...