Matches in SemOpenAlex for { <https://semopenalex.org/work/W3134865218> ?p ?o ?g. }
- W3134865218 abstract "Domain adaptation (DA) aims at transferring knowledge from a labeled source domain to an unlabeled target domain. Though many DA theories and algorithms have been proposed, most of them are tailored into classification settings and may fail in regression tasks, especially in the practical keypoint detection task. To tackle this difficult but significant task, we present a method of regressive domain adaptation (RegDA) for unsupervised keypoint detection. Inspired by the latest theoretical work, we first utilize an adversarial regressor to maximize the disparity on the target domain and train a feature generator to minimize this disparity. However, due to the high dimension of the output space, this regressor fails to detect samples that deviate from the support of the source. To overcome this problem, we propose two important ideas. First, based on our observation that the probability density of the output space is sparse, we introduce a spatial probability distribution to describe this sparsity and then use it to guide the learning of the adversarial regressor. Second, to alleviate the optimization difficulty in the high-dimensional space, we innovatively convert the minimax game in the adversarial training to the minimization of two opposite goals. Extensive experiments show that our method brings large improvement by 8% to 11% in terms of PCK on different datasets." @default.
- W3134865218 created "2021-03-15" @default.
- W3134865218 creator A5019241553 @default.
- W3134865218 creator A5034579434 @default.
- W3134865218 creator A5036144965 @default.
- W3134865218 creator A5046554988 @default.
- W3134865218 creator A5059586503 @default.
- W3134865218 creator A5074913703 @default.
- W3134865218 date "2021-03-10" @default.
- W3134865218 modified "2023-10-17" @default.
- W3134865218 title "Regressive Domain Adaptation for Unsupervised Keypoint Detection" @default.
- W3134865218 cites W1722318740 @default.
- W3134865218 cites W1731081199 @default.
- W3134865218 cites W2033547469 @default.
- W3134865218 cites W2034996255 @default.
- W3134865218 cites W2083544878 @default.
- W3134865218 cites W2101032778 @default.
- W3134865218 cites W2104094955 @default.
- W3134865218 cites W2110097068 @default.
- W3134865218 cites W2135533529 @default.
- W3134865218 cites W2136391815 @default.
- W3134865218 cites W2159291411 @default.
- W3134865218 cites W2194775991 @default.
- W3134865218 cites W2293220651 @default.
- W3134865218 cites W2307770531 @default.
- W3134865218 cites W2382036597 @default.
- W3134865218 cites W2576289912 @default.
- W3134865218 cites W2578797046 @default.
- W3134865218 cites W2618530766 @default.
- W3134865218 cites W2756050327 @default.
- W3134865218 cites W2796779902 @default.
- W3134865218 cites W2883939894 @default.
- W3134865218 cites W2892644985 @default.
- W3134865218 cites W2907715846 @default.
- W3134865218 cites W2916798096 @default.
- W3134865218 cites W2962687275 @default.
- W3134865218 cites W2962811204 @default.
- W3134865218 cites W2963094258 @default.
- W3134865218 cites W2963150697 @default.
- W3134865218 cites W2964105113 @default.
- W3134865218 cites W2964121744 @default.
- W3134865218 cites W2964297864 @default.
- W3134865218 cites W2964304707 @default.
- W3134865218 cites W3015865682 @default.
- W3134865218 doi "https://doi.org/10.48550/arxiv.2103.06175" @default.
- W3134865218 hasPublicationYear "2021" @default.
- W3134865218 type Work @default.
- W3134865218 sameAs 3134865218 @default.
- W3134865218 citedByCount "0" @default.
- W3134865218 crossrefType "posted-content" @default.
- W3134865218 hasAuthorship W3134865218A5019241553 @default.
- W3134865218 hasAuthorship W3134865218A5034579434 @default.
- W3134865218 hasAuthorship W3134865218A5036144965 @default.
- W3134865218 hasAuthorship W3134865218A5046554988 @default.
- W3134865218 hasAuthorship W3134865218A5059586503 @default.
- W3134865218 hasAuthorship W3134865218A5074913703 @default.
- W3134865218 hasBestOaLocation W31348652181 @default.
- W3134865218 hasConcept C119857082 @default.
- W3134865218 hasConcept C120665830 @default.
- W3134865218 hasConcept C121332964 @default.
- W3134865218 hasConcept C126255220 @default.
- W3134865218 hasConcept C134306372 @default.
- W3134865218 hasConcept C138885662 @default.
- W3134865218 hasConcept C139807058 @default.
- W3134865218 hasConcept C147764199 @default.
- W3134865218 hasConcept C149728462 @default.
- W3134865218 hasConcept C153180895 @default.
- W3134865218 hasConcept C154945302 @default.
- W3134865218 hasConcept C162324750 @default.
- W3134865218 hasConcept C163258240 @default.
- W3134865218 hasConcept C187736073 @default.
- W3134865218 hasConcept C199360897 @default.
- W3134865218 hasConcept C202444582 @default.
- W3134865218 hasConcept C2776401178 @default.
- W3134865218 hasConcept C2776434776 @default.
- W3134865218 hasConcept C2780451532 @default.
- W3134865218 hasConcept C2780992000 @default.
- W3134865218 hasConcept C33676613 @default.
- W3134865218 hasConcept C33923547 @default.
- W3134865218 hasConcept C36503486 @default.
- W3134865218 hasConcept C37736160 @default.
- W3134865218 hasConcept C41008148 @default.
- W3134865218 hasConcept C41895202 @default.
- W3134865218 hasConcept C62520636 @default.
- W3134865218 hasConcept C95623464 @default.
- W3134865218 hasConceptScore W3134865218C119857082 @default.
- W3134865218 hasConceptScore W3134865218C120665830 @default.
- W3134865218 hasConceptScore W3134865218C121332964 @default.
- W3134865218 hasConceptScore W3134865218C126255220 @default.
- W3134865218 hasConceptScore W3134865218C134306372 @default.
- W3134865218 hasConceptScore W3134865218C138885662 @default.
- W3134865218 hasConceptScore W3134865218C139807058 @default.
- W3134865218 hasConceptScore W3134865218C147764199 @default.
- W3134865218 hasConceptScore W3134865218C149728462 @default.
- W3134865218 hasConceptScore W3134865218C153180895 @default.
- W3134865218 hasConceptScore W3134865218C154945302 @default.
- W3134865218 hasConceptScore W3134865218C162324750 @default.
- W3134865218 hasConceptScore W3134865218C163258240 @default.
- W3134865218 hasConceptScore W3134865218C187736073 @default.
- W3134865218 hasConceptScore W3134865218C199360897 @default.