Matches in SemOpenAlex for { <https://semopenalex.org/work/W2605300166> ?p ?o ?g. }
- W2605300166 abstract "In contrast to the widely studied problem of recognizing an action given a complete sequence, action anticipation aims to identify the action from only partially available videos. As such, it is therefore key to the success of computer vision applications requiring to react as early as possible, such as autonomous navigation. In this paper, we propose a new action anticipation method that achieves high prediction accuracy even in the presence of a very small percentage of a video sequence. To this end, we develop a multi-stage LSTM architecture that leverages context-aware and action-aware features, and introduce a novel loss function that encourages the model to predict the correct class as early as possible. Our experiments on standard benchmark datasets evidence the benefits of our approach; We outperform the state-of-the-art action anticipation methods for early prediction by a relative increase in accuracy of 22.0% on JHMDB-21, 14.0% on UT-Interaction and 49.9% on UCF-101." @default.
- W2605300166 created "2017-04-14" @default.
- W2605300166 creator A5000886289 @default.
- W2605300166 creator A5003777910 @default.
- W2605300166 creator A5043470054 @default.
- W2605300166 creator A5049300388 @default.
- W2605300166 creator A5060335798 @default.
- W2605300166 creator A5090467618 @default.
- W2605300166 date "2017-10-01" @default.
- W2605300166 modified "2023-10-05" @default.
- W2605300166 title "Encouraging LSTMs to Anticipate Actions Very Early" @default.
- W2605300166 cites W10916678 @default.
- W2605300166 cites W1522734439 @default.
- W2605300166 cites W1930528368 @default.
- W2605300166 cites W1944615693 @default.
- W2605300166 cites W1945129080 @default.
- W2605300166 cites W1947481528 @default.
- W2605300166 cites W2016053056 @default.
- W2605300166 cites W2018068650 @default.
- W2605300166 cites W2020163092 @default.
- W2605300166 cites W2034014085 @default.
- W2605300166 cites W2089505226 @default.
- W2605300166 cites W2105101328 @default.
- W2605300166 cites W2147615062 @default.
- W2605300166 cites W2150354123 @default.
- W2605300166 cites W2167216530 @default.
- W2605300166 cites W2168328261 @default.
- W2605300166 cites W2295107390 @default.
- W2605300166 cites W2342662179 @default.
- W2605300166 cites W2422305492 @default.
- W2605300166 cites W2460134573 @default.
- W2605300166 cites W2461621749 @default.
- W2605300166 cites W2462996230 @default.
- W2605300166 cites W2465313502 @default.
- W2605300166 cites W2465503160 @default.
- W2605300166 cites W2471775118 @default.
- W2605300166 cites W2472970127 @default.
- W2605300166 cites W2533503513 @default.
- W2605300166 cites W2563717578 @default.
- W2605300166 cites W2963692464 @default.
- W2605300166 cites W2964072322 @default.
- W2605300166 doi "https://doi.org/10.1109/iccv.2017.39" @default.
- W2605300166 hasPublicationYear "2017" @default.
- W2605300166 type Work @default.
- W2605300166 sameAs 2605300166 @default.
- W2605300166 citedByCount "135" @default.
- W2605300166 countsByYear W26053001662017 @default.
- W2605300166 countsByYear W26053001662018 @default.
- W2605300166 countsByYear W26053001662019 @default.
- W2605300166 countsByYear W26053001662020 @default.
- W2605300166 countsByYear W26053001662021 @default.
- W2605300166 countsByYear W26053001662022 @default.
- W2605300166 countsByYear W26053001662023 @default.
- W2605300166 crossrefType "proceedings-article" @default.
- W2605300166 hasAuthorship W2605300166A5000886289 @default.
- W2605300166 hasAuthorship W2605300166A5003777910 @default.
- W2605300166 hasAuthorship W2605300166A5043470054 @default.
- W2605300166 hasAuthorship W2605300166A5049300388 @default.
- W2605300166 hasAuthorship W2605300166A5060335798 @default.
- W2605300166 hasAuthorship W2605300166A5090467618 @default.
- W2605300166 hasBestOaLocation W26053001662 @default.
- W2605300166 hasConcept C119857082 @default.
- W2605300166 hasConcept C121332964 @default.
- W2605300166 hasConcept C13280743 @default.
- W2605300166 hasConcept C151730666 @default.
- W2605300166 hasConcept C154945302 @default.
- W2605300166 hasConcept C176777502 @default.
- W2605300166 hasConcept C185798385 @default.
- W2605300166 hasConcept C205649164 @default.
- W2605300166 hasConcept C26517878 @default.
- W2605300166 hasConcept C2777212361 @default.
- W2605300166 hasConcept C2778112365 @default.
- W2605300166 hasConcept C2779343474 @default.
- W2605300166 hasConcept C2780791683 @default.
- W2605300166 hasConcept C2987834672 @default.
- W2605300166 hasConcept C38652104 @default.
- W2605300166 hasConcept C41008148 @default.
- W2605300166 hasConcept C54355233 @default.
- W2605300166 hasConcept C62520636 @default.
- W2605300166 hasConcept C86803240 @default.
- W2605300166 hasConceptScore W2605300166C119857082 @default.
- W2605300166 hasConceptScore W2605300166C121332964 @default.
- W2605300166 hasConceptScore W2605300166C13280743 @default.
- W2605300166 hasConceptScore W2605300166C151730666 @default.
- W2605300166 hasConceptScore W2605300166C154945302 @default.
- W2605300166 hasConceptScore W2605300166C176777502 @default.
- W2605300166 hasConceptScore W2605300166C185798385 @default.
- W2605300166 hasConceptScore W2605300166C205649164 @default.
- W2605300166 hasConceptScore W2605300166C26517878 @default.
- W2605300166 hasConceptScore W2605300166C2777212361 @default.
- W2605300166 hasConceptScore W2605300166C2778112365 @default.
- W2605300166 hasConceptScore W2605300166C2779343474 @default.
- W2605300166 hasConceptScore W2605300166C2780791683 @default.
- W2605300166 hasConceptScore W2605300166C2987834672 @default.
- W2605300166 hasConceptScore W2605300166C38652104 @default.
- W2605300166 hasConceptScore W2605300166C41008148 @default.
- W2605300166 hasConceptScore W2605300166C54355233 @default.
- W2605300166 hasConceptScore W2605300166C62520636 @default.
- W2605300166 hasConceptScore W2605300166C86803240 @default.
- W2605300166 hasLocation W26053001661 @default.