FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2164820723> ?p ?o ?g. }

• W2164820723 abstract "Over the last few years energy minimization has emerged as an indispensable tool in computer vision. The primary reason for this rising popularity has been the successes of efficient graph cut based minimization algorithms in solving many low level vision problems such as image segmentation, object reconstruction, image restoration and disparity estimation. The scale and form of computer vision problems introduce many challenges in energy minimization. In this dissertation, I will focus on some aspects of these problems. The first problem I address relates to the efficient and exact minimization of groups of similar functions which are known to be solvable in polynomial time. I will present a novel dynamic algorithm for minimizing such functions. This algorithm reuses computation from previous problem instances to solve new instances resulting in a substantial improvement in the running time. I will present the results of using this approach on the problems of interactive image segmentation, image segmentation in video, human pose estimation and segmentation, and measuring uncertainty of solutions obtained by minimizing energy functions. The second part of my dissertation will deal with the minimization of multilabel higher order functions which are np-hard to minimize. These functions are able to model interactions among groups of random variables and can be used to formulate many vision problems. The runtime complexity of commonly used algorithms for approximate energy minimization such as Max-product Belief Propagation or Tree-reweighted message passing grows exponentially with the clique size, which makes them inapplicable to problems with even moderate sized cliques. I will show how certain higher order energy functions can be minimized using the graph cut based expansion and swap move algorithms. This method is extremely efficient and is able to handle cliques involving thousands of variables. I will use these higher order energy functions to model the problems of object segmentation and recognition, and texture segmentation. The results of this approach will be compared with those obtained using conventional methods which model these problems using second order energy functions." @default.
• W2164820723 created "2016-06-24" @default.
• W2164820723 creator A5013834379 @default.
• W2164820723 date "2010-07-26" @default.
• W2164820723 modified "2023-09-26" @default.
• W2164820723 title "Minimizing Dynamic and Higher Order Energy Functions using Graph Cuts" @default.
• W2164820723 cites W10769007 @default.
• W2164820723 cites W123239406 @default.
• W2164820723 cites W1496918921 @default.
• W2164820723 cites W1498671329 @default.
• W2164820723 cites W149871427 @default.
• W2164820723 cites W1506851152 @default.
• W2164820723 cites W1516029338 @default.
• W2164820723 cites W1518641734 @default.
• W2164820723 cites W1528789833 @default.
• W2164820723 cites W1541388462 @default.
• W2164820723 cites W1546972969 @default.
• W2164820723 cites W1550397047 @default.
• W2164820723 cites W1576820654 @default.
• W2164820723 cites W1590899626 @default.
• W2164820723 cites W1643263348 @default.
• W2164820723 cites W1781337833 @default.
• W2164820723 cites W1785730614 @default.
• W2164820723 cites W1886770168 @default.
• W2164820723 cites W1963547452 @default.
• W2164820723 cites W1964119857 @default.
• W2164820723 cites W1964669037 @default.
• W2164820723 cites W1989339169 @default.
• W2164820723 cites W1992192543 @default.
• W2164820723 cites W1996540180 @default.
• W2164820723 cites W1999478155 @default.
• W2164820723 cites W2013469283 @default.
• W2164820723 cites W2029727948 @default.
• W2164820723 cites W2035575256 @default.
• W2164820723 cites W2050865875 @default.
• W2164820723 cites W2066509127 @default.
• W2164820723 cites W2067191022 @default.
• W2164820723 cites W2074078071 @default.
• W2164820723 cites W2078000663 @default.
• W2164820723 cites W2095844239 @default.
• W2164820723 cites W2097041931 @default.
• W2164820723 cites W2097713019 @default.
• W2164820723 cites W2098678088 @default.
• W2164820723 cites W2099835437 @default.
• W2164820723 cites W2100824854 @default.
• W2164820723 cites W2101309634 @default.
• W2164820723 cites W2102621940 @default.
• W2164820723 cites W2102625004 @default.
• W2164820723 cites W2102630415 @default.
• W2164820723 cites W2104020497 @default.
• W2164820723 cites W2104125540 @default.
• W2164820723 cites W2104974755 @default.
• W2164820723 cites W2105297815 @default.
• W2164820723 cites W2106110775 @default.
• W2164820723 cites W2107628931 @default.
• W2164820723 cites W2108619558 @default.
• W2164820723 cites W2112301665 @default.
• W2164820723 cites W2112556758 @default.
• W2164820723 cites W2113137767 @default.
• W2164820723 cites W2113206213 @default.
• W2164820723 cites W2114117477 @default.
• W2164820723 cites W2114614590 @default.
• W2164820723 cites W2115952485 @default.
• W2164820723 cites W2117435890 @default.
• W2164820723 cites W2120621946 @default.
• W2164820723 cites W2120689934 @default.
• W2164820723 cites W2120804060 @default.
• W2164820723 cites W2121845348 @default.
• W2164820723 cites W2121947440 @default.
• W2164820723 cites W2123282994 @default.
• W2164820723 cites W2124351162 @default.
• W2164820723 cites W2125489247 @default.
• W2164820723 cites W2126789858 @default.
• W2164820723 cites W2128059732 @default.
• W2164820723 cites W2128680590 @default.
• W2164820723 cites W2129823392 @default.
• W2164820723 cites W2129976136 @default.
• W2164820723 cites W2131432112 @default.
• W2164820723 cites W2131686571 @default.
• W2164820723 cites W2132055825 @default.
• W2164820723 cites W2134364100 @default.
• W2164820723 cites W2135165032 @default.
• W2164820723 cites W2135414191 @default.
• W2164820723 cites W2136224190 @default.
• W2164820723 cites W2137117160 @default.
• W2164820723 cites W2140502500 @default.
• W2164820723 cites W2140789035 @default.
• W2164820723 cites W2143075689 @default.
• W2164820723 cites W2143516773 @default.
• W2164820723 cites W2145108814 @default.
• W2164820723 cites W2146036075 @default.
• W2164820723 cites W2146293384 @default.
• W2164820723 cites W2147880316 @default.
• W2164820723 cites W2148018895 @default.
• W2164820723 cites W2149203124 @default.
• W2164820723 cites W2150095155 @default.
• W2164820723 cites W2151996626 @default.
• W2164820723 cites W2152858030 @default.
• W2164820723 cites W2152926413 @default.
• W2164820723 cites W2153980837 @default.

• next