FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2904324197> ?p ?o ?g. }

• W2904324197 endingPage "693" @default.
• W2904324197 startingPage "679" @default.
• W2904324197 abstract "In this paper, we propose an end-to-end reasoning-decision networks (RDN) approach for robust face alignment via policy gradient. Unlike the conventional coarse-to-fine approaches which likely lead to bias prediction due to poor initialization, our approach aims to learn a policy by leveraging raw pixels to reason a subset of shape candidates, sequentially making plausible decisions to remove outliers for robust initialization. To achieve this, we formulate face alignment as a Markov decision process by defining an agent, which typically interacts with a trajectory of states, actions, state transitions and rewards. The agent seeks an optimal shape searching policy over the whole shape space by maximizing a discounted sum of the received values. To further improve the alignment performance, we develop an LSTM-based value function to evaluate the shape quality. During the training procedure, we adjust the gradient of our value function in directions of the policy gradient. This prevents our training goal from being trapped into local optima entangled by both the pose deformations and appearance variations especially in unconstrained environments. Experimental results show that our proposed RDN consistently outperforms most state-of-the-art approaches on four widely-evaluated challenging datasets." @default.
• W2904324197 created "2018-12-22" @default.
• W2904324197 creator A5010357390 @default.
• W2904324197 creator A5018791285 @default.
• W2904324197 creator A5068171246 @default.
• W2904324197 creator A5088664989 @default.
• W2904324197 creator A5090079801 @default.
• W2904324197 date "2020-03-01" @default.
• W2904324197 modified "2023-10-17" @default.
• W2904324197 title "Learning Reasoning-Decision Networks for Robust Face Alignment" @default.
• W2904324197 cites W1599238028 @default.
• W2904324197 cites W1682276745 @default.
• W2904324197 cites W1795776638 @default.
• W2904324197 cites W1796263212 @default.
• W2904324197 cites W1915668717 @default.
• W2904324197 cites W1946919140 @default.
• W2904324197 cites W1976948919 @default.
• W2904324197 cites W1982368100 @default.
• W2904324197 cites W1998294030 @default.
• W2904324197 cites W2005264304 @default.
• W2904324197 cites W2012885984 @default.
• W2904324197 cites W2013564878 @default.
• W2904324197 cites W2032558548 @default.
• W2904324197 cites W2032618685 @default.
• W2904324197 cites W2058961190 @default.
• W2904324197 cites W2064675550 @default.
• W2904324197 cites W2076434944 @default.
• W2904324197 cites W2085983359 @default.
• W2904324197 cites W2087681821 @default.
• W2904324197 cites W2101866605 @default.
• W2904324197 cites W2111372597 @default.
• W2904324197 cites W2112695787 @default.
• W2904324197 cites W2119717200 @default.
• W2904324197 cites W2130563197 @default.
• W2904324197 cites W2145287260 @default.
• W2904324197 cites W2145339207 @default.
• W2904324197 cites W2152826865 @default.
• W2904324197 cites W2157285372 @default.
• W2904324197 cites W2166694921 @default.
• W2904324197 cites W2172532449 @default.
• W2904324197 cites W2214733281 @default.
• W2904324197 cites W2219124274 @default.
• W2904324197 cites W2221735899 @default.
• W2904324197 cites W2251810906 @default.
• W2904324197 cites W2257979135 @default.
• W2904324197 cites W2284800790 @default.
• W2904324197 cites W2307770531 @default.
• W2904324197 cites W2462523589 @default.
• W2904324197 cites W2465108587 @default.
• W2904324197 cites W2474575620 @default.
• W2904324197 cites W2506426544 @default.
• W2904324197 cites W2519753233 @default.
• W2904324197 cites W2520331172 @default.
• W2904324197 cites W2565729570 @default.
• W2904324197 cites W2580773671 @default.
• W2904324197 cites W2607151106 @default.
• W2904324197 cites W2738318237 @default.
• W2904324197 cites W2740020909 @default.
• W2904324197 cites W2740103755 @default.
• W2904324197 cites W2776507577 @default.
• W2904324197 cites W2798463715 @default.
• W2904324197 cites W2798644314 @default.
• W2904324197 cites W2798730128 @default.
• W2904324197 cites W2799171885 @default.
• W2904324197 cites W2919115771 @default.
• W2904324197 cites W2949376505 @default.
• W2904324197 cites W2952074561 @default.
• W2904324197 cites W2962890819 @default.
• W2904324197 cites W2963544488 @default.
• W2904324197 cites W2963789946 @default.
• W2904324197 cites W2963979855 @default.
• W2904324197 cites W2964145484 @default.
• W2904324197 cites W2964309795 @default.
• W2904324197 cites W3104792420 @default.
• W2904324197 doi "https://doi.org/10.1109/tpami.2018.2885298" @default.
• W2904324197 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/30530310" @default.
• W2904324197 hasPublicationYear "2020" @default.
• W2904324197 type Work @default.
• W2904324197 sameAs 2904324197 @default.
• W2904324197 citedByCount "13" @default.
• W2904324197 countsByYear W29043241972019 @default.
• W2904324197 countsByYear W29043241972020 @default.
• W2904324197 countsByYear W29043241972021 @default.
• W2904324197 countsByYear W29043241972022 @default.
• W2904324197 countsByYear W29043241972023 @default.
• W2904324197 crossrefType "journal-article" @default.
• W2904324197 hasAuthorship W2904324197A5010357390 @default.
• W2904324197 hasAuthorship W2904324197A5018791285 @default.
• W2904324197 hasAuthorship W2904324197A5068171246 @default.
• W2904324197 hasAuthorship W2904324197A5088664989 @default.
• W2904324197 hasAuthorship W2904324197A5090079801 @default.
• W2904324197 hasConcept C105795698 @default.
• W2904324197 hasConcept C106189395 @default.
• W2904324197 hasConcept C111919701 @default.
• W2904324197 hasConcept C114466953 @default.
• W2904324197 hasConcept C119857082 @default.
• W2904324197 hasConcept C121332964 @default.
• W2904324197 hasConcept C1276947 @default.

• next