Matches in SemOpenAlex for { <https://semopenalex.org/work/W4378647914> ?p ?o ?g. }
- W4378647914 endingPage "57136" @default.
- W4378647914 startingPage "57117" @default.
- W4378647914 abstract "Most learning-based methods require labelling the training data, which is time-consuming and gives rise to wrong labels. To address the labelling issues thoroughly, we propose an unsupervised learning framework to remove mismatches by maximizing the expected score of sample consensuses (MESAC). The proposed MESAC can train various permutation invariant networks (PINs) based on training data with no labels, and has three distinct merits: 1) the framework can train various PINs in an unsupervised mode such that these are immune to wrong labels; 2) the gradients of the expected score are explicitly calculated by a revised score-function estimator, which can avoid gradient explosion; 3) the distribution of matching probabilities is learned from the PIN and precisely modelled by a categorical distribution, which can decrease the sampling times and improve the computational efficiency accordingly. Experiments of testing datasets disclose that mean recall is increased by at most 77% when pure PINs are embedded in MESAC, and mean precision is also improved by 16%. Applications in pose recovery indicate that the success rates of MESAC-integrated PINs outperform the compared methods when training with neither matching labels nor ground truth epipolar geometry (EG) constraints, showing the great potential of MESAC in mismatch removal." @default.
- W4378647914 created "2023-05-30" @default.
- W4378647914 creator A5005890093 @default.
- W4378647914 creator A5009122391 @default.
- W4378647914 creator A5023705186 @default.
- W4378647914 creator A5027883397 @default.
- W4378647914 creator A5055546418 @default.
- W4378647914 creator A5066102428 @default.
- W4378647914 creator A5074212991 @default.
- W4378647914 date "2023-01-01" @default.
- W4378647914 modified "2023-09-25" @default.
- W4378647914 title "MESAC: Learning to Remove Mismatches via Maximizing the Expected Score of Sample Consensuses" @default.
- W4378647914 cites W1187712878 @default.
- W4378647914 cites W1578412938 @default.
- W4378647914 cites W1873846702 @default.
- W4378647914 cites W1985238052 @default.
- W4378647914 cites W2026375818 @default.
- W4378647914 cites W2046434485 @default.
- W4378647914 cites W2085261163 @default.
- W4378647914 cites W2099927035 @default.
- W4378647914 cites W2102481828 @default.
- W4378647914 cites W2106199912 @default.
- W4378647914 cites W2111073598 @default.
- W4378647914 cites W2114759747 @default.
- W4378647914 cites W2116598146 @default.
- W4378647914 cites W2130017587 @default.
- W4378647914 cites W2133192850 @default.
- W4378647914 cites W2151103935 @default.
- W4378647914 cites W2250384498 @default.
- W4378647914 cites W2290450005 @default.
- W4378647914 cites W2485734887 @default.
- W4378647914 cites W2610752411 @default.
- W4378647914 cites W2740578684 @default.
- W4378647914 cites W2754925132 @default.
- W4378647914 cites W2894971516 @default.
- W4378647914 cites W2948638726 @default.
- W4378647914 cites W2948859010 @default.
- W4378647914 cites W2955793676 @default.
- W4378647914 cites W2963674285 @default.
- W4378647914 cites W2982933447 @default.
- W4378647914 cites W2987672160 @default.
- W4378647914 cites W2990655570 @default.
- W4378647914 cites W3034373437 @default.
- W4378647914 cites W3035563186 @default.
- W4378647914 cites W3047057232 @default.
- W4378647914 cites W3091225957 @default.
- W4378647914 cites W3092233714 @default.
- W4378647914 cites W3096562082 @default.
- W4378647914 cites W3208916329 @default.
- W4378647914 cites W4244681853 @default.
- W4378647914 cites W4283379347 @default.
- W4378647914 cites W4283789017 @default.
- W4378647914 cites W4292313831 @default.
- W4378647914 cites W4312244264 @default.
- W4378647914 cites W4312322996 @default.
- W4378647914 cites W4319778255 @default.
- W4378647914 cites W57689863 @default.
- W4378647914 doi "https://doi.org/10.1109/access.2023.3280824" @default.
- W4378647914 hasPublicationYear "2023" @default.
- W4378647914 type Work @default.
- W4378647914 citedByCount "0" @default.
- W4378647914 crossrefType "journal-article" @default.
- W4378647914 hasAuthorship W4378647914A5005890093 @default.
- W4378647914 hasAuthorship W4378647914A5009122391 @default.
- W4378647914 hasAuthorship W4378647914A5023705186 @default.
- W4378647914 hasAuthorship W4378647914A5027883397 @default.
- W4378647914 hasAuthorship W4378647914A5055546418 @default.
- W4378647914 hasAuthorship W4378647914A5066102428 @default.
- W4378647914 hasAuthorship W4378647914A5074212991 @default.
- W4378647914 hasBestOaLocation W43786479141 @default.
- W4378647914 hasConcept C105795698 @default.
- W4378647914 hasConcept C107673813 @default.
- W4378647914 hasConcept C115961682 @default.
- W4378647914 hasConcept C119857082 @default.
- W4378647914 hasConcept C146849305 @default.
- W4378647914 hasConcept C153180895 @default.
- W4378647914 hasConcept C154945302 @default.
- W4378647914 hasConcept C160234255 @default.
- W4378647914 hasConcept C165064840 @default.
- W4378647914 hasConcept C185429906 @default.
- W4378647914 hasConcept C185592680 @default.
- W4378647914 hasConcept C190470478 @default.
- W4378647914 hasConcept C198531522 @default.
- W4378647914 hasConcept C23379248 @default.
- W4378647914 hasConcept C33923547 @default.
- W4378647914 hasConcept C37903108 @default.
- W4378647914 hasConcept C37914503 @default.
- W4378647914 hasConcept C41008148 @default.
- W4378647914 hasConcept C43617362 @default.
- W4378647914 hasConcept C46686674 @default.
- W4378647914 hasConcept C5274069 @default.
- W4378647914 hasConcept C65660741 @default.
- W4378647914 hasConcept C67926830 @default.
- W4378647914 hasConceptScore W4378647914C105795698 @default.
- W4378647914 hasConceptScore W4378647914C107673813 @default.
- W4378647914 hasConceptScore W4378647914C115961682 @default.
- W4378647914 hasConceptScore W4378647914C119857082 @default.
- W4378647914 hasConceptScore W4378647914C146849305 @default.