FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3109362868> ?p ?o ?g. }

• W3109362868 endingPage "120" @default.
• W3109362868 startingPage "104" @default.
• W3109362868 abstract "This paper tackles the challenging problem of estimating the intensity of Facial Action Units with few labeled images. Contrary to previous works, our method does not require to manually select key frames, and produces state-of-the-art results with as little as (2%) of annotated frames, which are randomly chosen. To this end, we propose a semi-supervised learning approach where a spatio-temporal model combining a feature extractor and a temporal module are learned in two stages. The first stage uses datasets of unlabeled videos to learn a strong spatio-temporal representation of facial behavior dynamics based on contrastive learning. To our knowledge we are the first to build upon this framework for modeling facial behavior in an unsupervised manner. The second stage uses another dataset of randomly chosen labeled frames to train a regressor on top of our spatio-temporal model for estimating the AU intensity. We show that although backpropagation through time is applied only with respect to the output of the network for extremely sparse and randomly chosen labeled frames, our model can be effectively trained to estimate AU intensity accurately, thanks to the unsupervised pre-training of the first stage. We experimentally validate that our method outperforms existing methods when working with as little as (2%) of randomly chosen data for both DISFA and BP4D datasets, without a careful choice of labeled frames, a time-consuming task still required in previous approaches." @default.
• W3109362868 created "2020-12-07" @default.
• W3109362868 creator A5024224610 @default.
• W3109362868 creator A5044266844 @default.
• W3109362868 creator A5057035871 @default.
• W3109362868 creator A5072822225 @default.
• W3109362868 date "2021-01-01" @default.
• W3109362868 modified "2023-09-27" @default.
• W3109362868 title "Semi-supervised Facial Action Unit Intensity Estimation with Contrastive Learning" @default.
• W3109362868 cites W1590099432 @default.
• W3109362868 cites W1616031638 @default.
• W3109362868 cites W1655469623 @default.
• W3109362868 cites W1677182931 @default.
• W3109362868 cites W2025768430 @default.
• W3109362868 cites W2045472600 @default.
• W3109362868 cites W2051297709 @default.
• W3109362868 cites W2141403362 @default.
• W3109362868 cites W2142575466 @default.
• W3109362868 cites W2143334932 @default.
• W3109362868 cites W2150355110 @default.
• W3109362868 cites W2157331557 @default.
• W3109362868 cites W2194775991 @default.
• W3109362868 cites W2280620570 @default.
• W3109362868 cites W2326925005 @default.
• W3109362868 cites W2421475762 @default.
• W3109362868 cites W2469434562 @default.
• W3109362868 cites W2474166357 @default.
• W3109362868 cites W2487442924 @default.
• W3109362868 cites W2589142773 @default.
• W3109362868 cites W2612215259 @default.
• W3109362868 cites W2655404332 @default.
• W3109362868 cites W2713788831 @default.
• W3109362868 cites W2756460808 @default.
• W3109362868 cites W2798764454 @default.
• W3109362868 cites W2798833386 @default.
• W3109362868 cites W2798991696 @default.
• W3109362868 cites W2799151537 @default.
• W3109362868 cites W2883725317 @default.
• W3109362868 cites W2943929354 @default.
• W3109362868 cites W2945559452 @default.
• W3109362868 cites W2948303854 @default.
• W3109362868 cites W2963292431 @default.
• W3109362868 cites W2963307811 @default.
• W3109362868 cites W2963713173 @default.
• W3109362868 cites W2963875208 @default.
• W3109362868 cites W2964322530 @default.
• W3109362868 cites W2979579363 @default.
• W3109362868 cites W2981595954 @default.
• W3109362868 cites W2995659499 @default.
• W3109362868 cites W3009809768 @default.
• W3109362868 cites W3010874390 @default.
• W3109362868 cites W3035524453 @default.
• W3109362868 cites W3108655343 @default.
• W3109362868 cites W3126750668 @default.
• W3109362868 cites W3130864507 @default.
• W3109362868 cites W343636949 @default.
• W3109362868 doi "https://doi.org/10.1007/978-3-030-69541-5_7" @default.
• W3109362868 hasPublicationYear "2021" @default.
• W3109362868 type Work @default.
• W3109362868 sameAs 3109362868 @default.
• W3109362868 citedByCount "0" @default.
• W3109362868 crossrefType "book-chapter" @default.
• W3109362868 hasAuthorship W3109362868A5024224610 @default.
• W3109362868 hasAuthorship W3109362868A5044266844 @default.
• W3109362868 hasAuthorship W3109362868A5057035871 @default.
• W3109362868 hasAuthorship W3109362868A5072822225 @default.
• W3109362868 hasBestOaLocation W31093628682 @default.
• W3109362868 hasConcept C117978034 @default.
• W3109362868 hasConcept C119857082 @default.
• W3109362868 hasConcept C127413603 @default.
• W3109362868 hasConcept C138885662 @default.
• W3109362868 hasConcept C153180895 @default.
• W3109362868 hasConcept C154945302 @default.
• W3109362868 hasConcept C162324750 @default.
• W3109362868 hasConcept C17744445 @default.
• W3109362868 hasConcept C187736073 @default.
• W3109362868 hasConcept C199539241 @default.
• W3109362868 hasConcept C21880701 @default.
• W3109362868 hasConcept C2776359362 @default.
• W3109362868 hasConcept C2776401178 @default.
• W3109362868 hasConcept C2780451532 @default.
• W3109362868 hasConcept C41008148 @default.
• W3109362868 hasConcept C41895202 @default.
• W3109362868 hasConcept C50644808 @default.
• W3109362868 hasConcept C59404180 @default.
• W3109362868 hasConcept C94625758 @default.
• W3109362868 hasConceptScore W3109362868C117978034 @default.
• W3109362868 hasConceptScore W3109362868C119857082 @default.
• W3109362868 hasConceptScore W3109362868C127413603 @default.
• W3109362868 hasConceptScore W3109362868C138885662 @default.
• W3109362868 hasConceptScore W3109362868C153180895 @default.
• W3109362868 hasConceptScore W3109362868C154945302 @default.
• W3109362868 hasConceptScore W3109362868C162324750 @default.
• W3109362868 hasConceptScore W3109362868C17744445 @default.
• W3109362868 hasConceptScore W3109362868C187736073 @default.
• W3109362868 hasConceptScore W3109362868C199539241 @default.
• W3109362868 hasConceptScore W3109362868C21880701 @default.
• W3109362868 hasConceptScore W3109362868C2776359362 @default.

• next