Matches in SemOpenAlex for { <https://semopenalex.org/work/W2945480377> ?p ?o ?g. }
- W2945480377 abstract "Partially annotated clips contain rich temporal contexts that can complement the sparse key frame annotations in providing supervision for model training. We present a novel paradigm called Temporally-Adaptive Features (TAF) learning that can utilize such data to learn better single frame models. By imposing distinct temporal change rate constraints on different factors in the model, TAF enables learning from unlabeled frames using context to enhance model accuracy. TAF generalizes slow feature learning and we present much stronger empirical evidence than prior works, showing convincing gains for the challenging semantic segmentation task over a variety of architecture designs and on two popular datasets. TAF can be interpreted as an implicit label augmentation method but is a more principled formulation compared to existing explicit augmentation techniques. Our work thus connects two promising methods that utilize partially annotated clips for single frame model training and can inspire future explorations in this direction." @default.
- W2945480377 created "2019-05-29" @default.
- W2945480377 creator A5042006378 @default.
- W2945480377 creator A5059310658 @default.
- W2945480377 creator A5070442175 @default.
- W2945480377 date "2019-05-24" @default.
- W2945480377 modified "2023-09-23" @default.
- W2945480377 title "Implicit Label Augmentation on Partially Annotated Clips via Temporally-Adaptive Features Learning." @default.
- W2945480377 cites W1599058448 @default.
- W2945480377 cites W1836533770 @default.
- W2945480377 cites W1903029394 @default.
- W2945480377 cites W1913356549 @default.
- W2945480377 cites W2004915807 @default.
- W2945480377 cites W2074772891 @default.
- W2945480377 cites W2095705004 @default.
- W2945480377 cites W2099471712 @default.
- W2945480377 cites W2113325037 @default.
- W2945480377 cites W2116435618 @default.
- W2945480377 cites W2117539524 @default.
- W2945480377 cites W2133665775 @default.
- W2945480377 cites W2144513243 @default.
- W2945480377 cites W2146444479 @default.
- W2945480377 cites W219040644 @default.
- W2945480377 cites W2194775991 @default.
- W2945480377 cites W2331128040 @default.
- W2945480377 cites W2340897893 @default.
- W2945480377 cites W2412782625 @default.
- W2945480377 cites W2470142083 @default.
- W2945480377 cites W2487442924 @default.
- W2945480377 cites W2520707372 @default.
- W2945480377 cites W2528342806 @default.
- W2945480377 cites W2559156603 @default.
- W2945480377 cites W2560023338 @default.
- W2945480377 cites W2561585794 @default.
- W2945480377 cites W2562457735 @default.
- W2945480377 cites W2565639579 @default.
- W2945480377 cites W2575671312 @default.
- W2945480377 cites W2594156432 @default.
- W2945480377 cites W2595272013 @default.
- W2945480377 cites W2609883120 @default.
- W2945480377 cites W2613718673 @default.
- W2945480377 cites W2630837129 @default.
- W2945480377 cites W2746314669 @default.
- W2945480377 cites W2768184749 @default.
- W2945480377 cites W2768591600 @default.
- W2945480377 cites W2770173563 @default.
- W2945480377 cites W2777795072 @default.
- W2945480377 cites W2793693263 @default.
- W2945480377 cites W2795912842 @default.
- W2945480377 cites W2798040152 @default.
- W2945480377 cites W2804047946 @default.
- W2945480377 cites W2890949887 @default.
- W2945480377 cites W2898091194 @default.
- W2945480377 cites W2902918442 @default.
- W2945480377 cites W2949634581 @default.
- W2945480377 cites W2955792047 @default.
- W2945480377 cites W2962785568 @default.
- W2945480377 cites W2962815021 @default.
- W2945480377 cites W2962824366 @default.
- W2945480377 cites W2962850830 @default.
- W2945480377 cites W2962889061 @default.
- W2945480377 cites W2962920946 @default.
- W2945480377 cites W2962958090 @default.
- W2945480377 cites W2963019093 @default.
- W2945480377 cites W2963150697 @default.
- W2945480377 cites W2963163009 @default.
- W2945480377 cites W2963351448 @default.
- W2945480377 cites W2963488291 @default.
- W2945480377 cites W2963583471 @default.
- W2945480377 cites W2963709863 @default.
- W2945480377 cites W2963749571 @default.
- W2945480377 cites W2963790522 @default.
- W2945480377 cites W2963840672 @default.
- W2945480377 cites W2963906250 @default.
- W2945480377 cites W2963919294 @default.
- W2945480377 cites W2963942586 @default.
- W2945480377 cites W2964236837 @default.
- W2945480377 cites W2964250774 @default.
- W2945480377 cites W2964309882 @default.
- W2945480377 cites W3012573144 @default.
- W2945480377 cites W3106250896 @default.
- W2945480377 cites W4919037 @default.
- W2945480377 cites W2963543962 @default.
- W2945480377 hasPublicationYear "2019" @default.
- W2945480377 type Work @default.
- W2945480377 sameAs 2945480377 @default.
- W2945480377 citedByCount "0" @default.
- W2945480377 crossrefType "posted-content" @default.
- W2945480377 hasAuthorship W2945480377A5042006378 @default.
- W2945480377 hasAuthorship W2945480377A5059310658 @default.
- W2945480377 hasAuthorship W2945480377A5070442175 @default.
- W2945480377 hasConcept C104317684 @default.
- W2945480377 hasConcept C112313634 @default.
- W2945480377 hasConcept C119857082 @default.
- W2945480377 hasConcept C126042441 @default.
- W2945480377 hasConcept C127716648 @default.
- W2945480377 hasConcept C138885662 @default.
- W2945480377 hasConcept C151730666 @default.
- W2945480377 hasConcept C154945302 @default.
- W2945480377 hasConcept C162324750 @default.