FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2956087352> ?p ?o ?g. }

• W2956087352 abstract "We introduce a method for learning landmark detectors from unlabelled video frames and unpaired labels. This allows us to learn a detector from a large collection of raw videos given only a few example annotations harvested from existing data or motion capture. We achieve this by formulating the landmark detection task as one of image translation, learning to map an image of the object to an image of its landmarks, represented as a skeleton. The advantage is that this translation problem can then be tackled by CycleGAN. However, we show that a naive application of CycleGAN confounds appearance and pose information, with suboptimal keypoint detection performance. We solve this problem by introducing an analytical and differentiable renderer for the skeleton image so that no appearance information can be leaked in the skeleton. Then, since cycle consistency requires to reconstruct the input image from the skeleton, we supply the appearance information thus removed by conditioning the generator with a second image of the same object (e.g. another frame from a video). Furthermore, while CycleGAN uses two cycle consistency constraints, we show that the second one is detrimental in this application and we discard it, significantly simplifying the model. We show that these modifications improve the quality of the learned detector leading to state-of-the-art unsupervised landmark detection performance in a number of challenging human pose and facial landmark detection benchmarks." @default.
• W2956087352 created "2019-07-12" @default.
• W2956087352 creator A5056977175 @default.
• W2956087352 creator A5060511349 @default.
• W2956087352 creator A5082195057 @default.
• W2956087352 creator A5086192896 @default.
• W2956087352 date "2019-07-03" @default.
• W2956087352 modified "2023-09-27" @default.
• W2956087352 title "Learning Landmarks from Unaligned Data using Image Translation" @default.
• W2956087352 cites W1795776638 @default.
• W2956087352 cites W1925384340 @default.
• W2956087352 cites W1943191679 @default.
• W2956087352 cites W1960706641 @default.
• W2956087352 cites W1996478295 @default.
• W2956087352 cites W1997691213 @default.
• W2956087352 cites W1998294030 @default.
• W2956087352 cites W2030536784 @default.
• W2956087352 cites W2032558548 @default.
• W2956087352 cites W2047508432 @default.
• W2956087352 cites W2080873731 @default.
• W2956087352 cites W2097151019 @default.
• W2956087352 cites W2099471712 @default.
• W2956087352 cites W2101032778 @default.
• W2956087352 cites W2103015390 @default.
• W2956087352 cites W2113325037 @default.
• W2956087352 cites W2115252128 @default.
• W2956087352 cites W2129210471 @default.
• W2956087352 cites W2131263044 @default.
• W2956087352 cites W2137591992 @default.
• W2956087352 cites W2155759509 @default.
• W2956087352 cites W2174722029 @default.
• W2956087352 cites W2214409633 @default.
• W2956087352 cites W2307770531 @default.
• W2956087352 cites W2308529009 @default.
• W2956087352 cites W2331128040 @default.
• W2956087352 cites W2412002662 @default.
• W2956087352 cites W2475287302 @default.
• W2956087352 cites W2502312327 @default.
• W2956087352 cites W2504108613 @default.
• W2956087352 cites W2518965973 @default.
• W2956087352 cites W2519753233 @default.
• W2956087352 cites W2558661413 @default.
• W2956087352 cites W2593414223 @default.
• W2956087352 cites W2593768305 @default.
• W2956087352 cites W2604233003 @default.
• W2956087352 cites W2607738331 @default.
• W2956087352 cites W2770121394 @default.
• W2956087352 cites W2774208477 @default.
• W2956087352 cites W2777262900 @default.
• W2956087352 cites W2795089319 @default.
• W2956087352 cites W2797184202 @default.
• W2956087352 cites W2803936827 @default.
• W2956087352 cites W2887997593 @default.
• W2956087352 cites W2890967717 @default.
• W2956087352 cites W2895439318 @default.
• W2956087352 cites W2951707615 @default.
• W2956087352 cites W2952422028 @default.
• W2956087352 cites W2962793481 @default.
• W2956087352 cites W2962808524 @default.
• W2956087352 cites W2962981304 @default.
• W2956087352 cites W2963073614 @default.
• W2956087352 cites W2963174698 @default.
• W2956087352 cites W2963419579 @default.
• W2956087352 cites W2963420272 @default.
• W2956087352 cites W2963448913 @default.
• W2956087352 cites W2963474899 @default.
• W2956087352 cites W2963590054 @default.
• W2956087352 cites W2963784072 @default.
• W2956087352 cites W2963823554 @default.
• W2956087352 cites W2963826681 @default.
• W2956087352 cites W2963906250 @default.
• W2956087352 cites W2963995996 @default.
• W2956087352 cites W2964304707 @default.
• W2956087352 cites W3022933286 @default.
• W2956087352 cites W602397586 @default.
• W2956087352 cites W2345945060 @default.
• W2956087352 hasPublicationYear "2019" @default.
• W2956087352 type Work @default.
• W2956087352 sameAs 2956087352 @default.
• W2956087352 citedByCount "6" @default.
• W2956087352 countsByYear W29560873522020 @default.
• W2956087352 countsByYear W29560873522021 @default.
• W2956087352 crossrefType "posted-content" @default.
• W2956087352 hasAuthorship W2956087352A5056977175 @default.
• W2956087352 hasAuthorship W2956087352A5060511349 @default.
• W2956087352 hasAuthorship W2956087352A5082195057 @default.
• W2956087352 hasAuthorship W2956087352A5086192896 @default.
• W2956087352 hasConcept C104317684 @default.
• W2956087352 hasConcept C105580179 @default.
• W2956087352 hasConcept C115961682 @default.
• W2956087352 hasConcept C121332964 @default.
• W2956087352 hasConcept C126042441 @default.
• W2956087352 hasConcept C149364088 @default.
• W2956087352 hasConcept C153180895 @default.
• W2956087352 hasConcept C154945302 @default.
• W2956087352 hasConcept C162324750 @default.
• W2956087352 hasConcept C163258240 @default.
• W2956087352 hasConcept C185592680 @default.
• W2956087352 hasConcept C187736073 @default.
• W2956087352 hasConcept C18969341 @default.

• next