Matches in SemOpenAlex for { <https://semopenalex.org/work/W3014983628> ?p ?o ?g. }
- W3014983628 abstract "Training deep neural networks to estimate the viewpoint of objects requires large labeled training datasets. However, manually labeling viewpoints is notoriously hard, error-prone, and time-consuming. On the other hand, it is relatively easy to mine many unlabelled images of an object category from the internet, e.g., of cars or faces. We seek to answer the research question of whether such unlabeled collections of in-the-wild images can be successfully utilized to train viewpoint estimation networks for general object categories purely via self-supervision. Self-supervision here refers to the fact that the only true supervisory signal that the network has is the input image itself. We propose a novel learning framework which incorporates an analysis-by-synthesis paradigm to reconstruct images in a viewpoint aware manner with a generative network, along with symmetry and adversarial constraints to successfully supervise our viewpoint estimation network. We show that our approach performs competitively to fully-supervised approaches for several object categories like human faces, cars, buses, and trains. Our work opens up further research in self-supervised viewpoint learning and serves as a robust baseline for it. We open-source our code at this https URL." @default.
- W3014983628 created "2020-04-10" @default.
- W3014983628 creator A5021623206 @default.
- W3014983628 creator A5021827101 @default.
- W3014983628 creator A5034290170 @default.
- W3014983628 creator A5049815485 @default.
- W3014983628 creator A5056503617 @default.
- W3014983628 creator A5067824015 @default.
- W3014983628 creator A5085870393 @default.
- W3014983628 date "2020-04-03" @default.
- W3014983628 modified "2023-09-26" @default.
- W3014983628 title "Self-Supervised Viewpoint Learning From Image Collections" @default.
- W3014983628 cites W1591870335 @default.
- W3014983628 cites W1691728462 @default.
- W3014983628 cites W1896788142 @default.
- W3014983628 cites W1949483711 @default.
- W3014983628 cites W1958236864 @default.
- W3014983628 cites W1976948919 @default.
- W3014983628 cites W1991264156 @default.
- W3014983628 cites W1991544872 @default.
- W3014983628 cites W2047875689 @default.
- W3014983628 cites W2058961190 @default.
- W3014983628 cites W2087681821 @default.
- W3014983628 cites W2099471712 @default.
- W3014983628 cites W2108598243 @default.
- W3014983628 cites W2147336195 @default.
- W3014983628 cites W2154833362 @default.
- W3014983628 cites W2237250383 @default.
- W3014983628 cites W2265959009 @default.
- W3014983628 cites W2518803647 @default.
- W3014983628 cites W2519379752 @default.
- W3014983628 cites W2558661413 @default.
- W3014983628 cites W2560544142 @default.
- W3014983628 cites W2572730214 @default.
- W3014983628 cites W2589255576 @default.
- W3014983628 cites W2600447016 @default.
- W3014983628 cites W2603777577 @default.
- W3014983628 cites W2737644856 @default.
- W3014983628 cites W2739748921 @default.
- W3014983628 cites W2771328060 @default.
- W3014983628 cites W2795096917 @default.
- W3014983628 cites W2796822548 @default.
- W3014983628 cites W2807744618 @default.
- W3014983628 cites W2883725317 @default.
- W3014983628 cites W2889468636 @default.
- W3014983628 cites W2889582485 @default.
- W3014983628 cites W2890967717 @default.
- W3014983628 cites W2899771611 @default.
- W3014983628 cites W2903200123 @default.
- W3014983628 cites W2929298487 @default.
- W3014983628 cites W2951880955 @default.
- W3014983628 cites W2957744218 @default.
- W3014983628 cites W2962688642 @default.
- W3014983628 cites W2962748819 @default.
- W3014983628 cites W2962770929 @default.
- W3014983628 cites W2962784289 @default.
- W3014983628 cites W2962835968 @default.
- W3014983628 cites W2962887041 @default.
- W3014983628 cites W2962967409 @default.
- W3014983628 cites W2962981304 @default.
- W3014983628 cites W2963002018 @default.
- W3014983628 cites W2963026643 @default.
- W3014983628 cites W2963072537 @default.
- W3014983628 cites W2963188159 @default.
- W3014983628 cites W2963226019 @default.
- W3014983628 cites W2963338719 @default.
- W3014983628 cites W2963342110 @default.
- W3014983628 cites W2963408523 @default.
- W3014983628 cites W2963419579 @default.
- W3014983628 cites W2963475767 @default.
- W3014983628 cites W2963527086 @default.
- W3014983628 cites W2963598268 @default.
- W3014983628 cites W2963706553 @default.
- W3014983628 cites W2963842958 @default.
- W3014983628 cites W2963850211 @default.
- W3014983628 cites W2964053173 @default.
- W3014983628 cites W2964073328 @default.
- W3014983628 cites W2964121744 @default.
- W3014983628 cites W2964171387 @default.
- W3014983628 cites W2965412140 @default.
- W3014983628 cites W2966661 @default.
- W3014983628 cites W2968257580 @default.
- W3014983628 cites W2969485315 @default.
- W3014983628 cites W2982024234 @default.
- W3014983628 cites W2982717194 @default.
- W3014983628 cites W2982966540 @default.
- W3014983628 cites W2990173985 @default.
- W3014983628 cites W3000817459 @default.
- W3014983628 cites W3101998545 @default.
- W3014983628 hasPublicationYear "2020" @default.
- W3014983628 type Work @default.
- W3014983628 sameAs 3014983628 @default.
- W3014983628 citedByCount "1" @default.
- W3014983628 countsByYear W30149836282020 @default.
- W3014983628 crossrefType "posted-content" @default.
- W3014983628 hasAuthorship W3014983628A5021623206 @default.
- W3014983628 hasAuthorship W3014983628A5021827101 @default.
- W3014983628 hasAuthorship W3014983628A5034290170 @default.
- W3014983628 hasAuthorship W3014983628A5049815485 @default.
- W3014983628 hasAuthorship W3014983628A5056503617 @default.