FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4362496215> ?p ?o ?g. }

• W4362496215 endingPage "2119" @default.
• W4362496215 startingPage "2107" @default.
• W4362496215 abstract "Domain generalizable person re-identification (DG ReID) is a challenging problem, because the trained model is often not generalizable to unseen target domains with different distribution from the source training domains. Data augmentation has been verified to be beneficial for better exploiting the source data to improve the model generalization. However, existing approaches primarily rely on pixel-level image generation that requires designing and training an extra generation network, which is extremely complex and provides limited diversity of augmented data. In this paper, we propose a simple yet effective feature based augmentation technique, named Style-uncertainty Augmentation (SuA). The main idea of SuA is to randomize the style of training data by perturbing the instance style with Gaussian noise during training process to increase the training domain diversity. And to better generalize knowledge across these augmented domains, we propose a progressive learning to learn strategy named Self-paced Meta Learning (SpML) that extends the conventional one-stage meta learning to multi-stage training process. The rationality is to gradually improve the model generalization ability to unseen target domains by simulating the mechanism of human learning. Furthermore, conventional person Re-ID loss functions are unable to leverage the valuable <italic xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>domain</i> information to improve the model generalization. So we further propose a distance-graph alignment loss that aligns the feature relationship distribution among domains to facilitate the network to explore domain-invariant representations of images. Extensive experiments on four large-scale benchmarks demonstrate that our SuA-SpML achieves state-of-the-art generalization to unseen domains for person ReID." @default.
• W4362496215 created "2023-04-05" @default.
• W4362496215 creator A5004565086 @default.
• W4362496215 creator A5012752139 @default.
• W4362496215 creator A5045284321 @default.
• W4362496215 creator A5060178337 @default.
• W4362496215 date "2023-01-01" @default.
• W4362496215 modified "2023-09-26" @default.
• W4362496215 title "Style Uncertainty Based Self-Paced Meta Learning for Generalizable Person Re-Identification" @default.
• W4362496215 cites W1915485278 @default.
• W4362496215 cites W1982925187 @default.
• W4362496215 cites W2108598243 @default.
• W4362496215 cites W2194775991 @default.
• W4362496215 cites W2204750386 @default.
• W4362496215 cites W2342611082 @default.
• W4362496215 cites W2511791013 @default.
• W4362496215 cites W2572730214 @default.
• W4362496215 cites W2584637367 @default.
• W4362496215 cites W2585635281 @default.
• W4362496215 cites W2603777577 @default.
• W4362496215 cites W2606377603 @default.
• W4362496215 cites W2778652957 @default.
• W4362496215 cites W2798658180 @default.
• W4362496215 cites W2884366600 @default.
• W4362496215 cites W2896016251 @default.
• W4362496215 cites W2898047322 @default.
• W4362496215 cites W2904427185 @default.
• W4362496215 cites W2947319827 @default.
• W4362496215 cites W2957037461 @default.
• W4362496215 cites W2958360136 @default.
• W4362496215 cites W2962793481 @default.
• W4362496215 cites W2962859295 @default.
• W4362496215 cites W2962926870 @default.
• W4362496215 cites W2963000559 @default.
• W4362496215 cites W2963043696 @default.
• W4362496215 cites W2963047834 @default.
• W4362496215 cites W2963049565 @default.
• W4362496215 cites W2963289251 @default.
• W4362496215 cites W2963557071 @default.
• W4362496215 cites W2963842104 @default.
• W4362496215 cites W2963975998 @default.
• W4362496215 cites W2979931389 @default.
• W4362496215 cites W2979938149 @default.
• W4362496215 cites W2981368934 @default.
• W4362496215 cites W2988852559 @default.
• W4362496215 cites W2989923292 @default.
• W4362496215 cites W2990827756 @default.
• W4362496215 cites W2998712190 @default.
• W4362496215 cites W3009761962 @default.
• W4362496215 cites W3034727830 @default.
• W4362496215 cites W3035070480 @default.
• W4362496215 cites W3039204154 @default.
• W4362496215 cites W3095467524 @default.
• W4362496215 cites W3095799614 @default.
• W4362496215 cites W3096285474 @default.
• W4362496215 cites W3104599541 @default.
• W4362496215 cites W3107836944 @default.
• W4362496215 cites W3109831211 @default.
• W4362496215 cites W3113148327 @default.
• W4362496215 cites W3113448926 @default.
• W4362496215 cites W3120798330 @default.
• W4362496215 cites W3166601111 @default.
• W4362496215 cites W3173909755 @default.
• W4362496215 cites W3174337559 @default.
• W4362496215 cites W3175215201 @default.
• W4362496215 cites W3175956495 @default.
• W4362496215 cites W3177182473 @default.
• W4362496215 cites W3179233662 @default.
• W4362496215 cites W3187816944 @default.
• W4362496215 cites W3195923711 @default.
• W4362496215 cites W3201074629 @default.
• W4362496215 cites W3201928165 @default.
• W4362496215 cites W3202100376 @default.
• W4362496215 cites W4214611271 @default.
• W4362496215 doi "https://doi.org/10.1109/tip.2023.3263112" @default.
• W4362496215 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37023142" @default.
• W4362496215 hasPublicationYear "2023" @default.
• W4362496215 type Work @default.
• W4362496215 citedByCount "1" @default.
• W4362496215 countsByYear W43624962152023 @default.
• W4362496215 crossrefType "journal-article" @default.
• W4362496215 hasAuthorship W4362496215A5004565086 @default.
• W4362496215 hasAuthorship W4362496215A5012752139 @default.
• W4362496215 hasAuthorship W4362496215A5045284321 @default.
• W4362496215 hasAuthorship W4362496215A5060178337 @default.
• W4362496215 hasConcept C108583219 @default.
• W4362496215 hasConcept C119857082 @default.
• W4362496215 hasConcept C134306372 @default.
• W4362496215 hasConcept C153083717 @default.
• W4362496215 hasConcept C154945302 @default.
• W4362496215 hasConcept C162324750 @default.
• W4362496215 hasConcept C177148314 @default.
• W4362496215 hasConcept C187736073 @default.
• W4362496215 hasConcept C197352929 @default.
• W4362496215 hasConcept C2778827112 @default.
• W4362496215 hasConcept C2780451532 @default.
• W4362496215 hasConcept C28006648 @default.
• W4362496215 hasConcept C33923547 @default.

• next