Matches in SemOpenAlex for { <https://semopenalex.org/work/W4288797768> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W4288797768 abstract "Labeling pixel-level masks for fine-grained semantic segmentation tasks, e.g. human parsing, remains a challenging task. The ambiguous boundary between different semantic parts and those categories with similar appearance usually are confusing, leading to unexpected noises in ground truth masks. To tackle the problem of learning with label noises, this work introduces a purification strategy, called Self-Correction for Human Parsing (SCHP), to progressively promote the reliability of the supervised labels as well as the learned models. In particular, starting from a model trained with inaccurate annotations as initialization, we design a cyclically learning scheduler to infer more reliable pseudo-masks by iteratively aggregating the current learned model with the former optimal one in an online manner. Besides, those correspondingly corrected labels can in turn to further boost the model performance. In this way, the models and the labels will reciprocally become more robust and accurate during the self-correction learning cycles. Benefiting from the superiority of SCHP, we achieve the best performance on two popular single-person human parsing benchmarks, including LIP and Pascal-Person-Part datasets. Our overall system ranks 1st in CVPR2019 LIP Challenge. Code is available at https://github.com/PeikeLi/Self-Correction-Human-Parsing." @default.
- W4288797768 created "2022-07-30" @default.
- W4288797768 creator A5032941294 @default.
- W4288797768 creator A5047384983 @default.
- W4288797768 creator A5057243649 @default.
- W4288797768 creator A5087043856 @default.
- W4288797768 date "2019-10-22" @default.
- W4288797768 modified "2023-09-29" @default.
- W4288797768 title "Self-Correction for Human Parsing" @default.
- W4288797768 doi "https://doi.org/10.48550/arxiv.1910.09777" @default.
- W4288797768 hasPublicationYear "2019" @default.
- W4288797768 type Work @default.
- W4288797768 citedByCount "0" @default.
- W4288797768 crossrefType "posted-content" @default.
- W4288797768 hasAuthorship W4288797768A5032941294 @default.
- W4288797768 hasAuthorship W4288797768A5047384983 @default.
- W4288797768 hasAuthorship W4288797768A5057243649 @default.
- W4288797768 hasAuthorship W4288797768A5087043856 @default.
- W4288797768 hasBestOaLocation W42887977681 @default.
- W4288797768 hasConcept C114466953 @default.
- W4288797768 hasConcept C119857082 @default.
- W4288797768 hasConcept C146849305 @default.
- W4288797768 hasConcept C149635348 @default.
- W4288797768 hasConcept C154945302 @default.
- W4288797768 hasConcept C162324750 @default.
- W4288797768 hasConcept C177264268 @default.
- W4288797768 hasConcept C186644900 @default.
- W4288797768 hasConcept C187736073 @default.
- W4288797768 hasConcept C199360897 @default.
- W4288797768 hasConcept C204321447 @default.
- W4288797768 hasConcept C2776760102 @default.
- W4288797768 hasConcept C2780451532 @default.
- W4288797768 hasConcept C2780513914 @default.
- W4288797768 hasConcept C41008148 @default.
- W4288797768 hasConcept C75608658 @default.
- W4288797768 hasConcept C89600930 @default.
- W4288797768 hasConceptScore W4288797768C114466953 @default.
- W4288797768 hasConceptScore W4288797768C119857082 @default.
- W4288797768 hasConceptScore W4288797768C146849305 @default.
- W4288797768 hasConceptScore W4288797768C149635348 @default.
- W4288797768 hasConceptScore W4288797768C154945302 @default.
- W4288797768 hasConceptScore W4288797768C162324750 @default.
- W4288797768 hasConceptScore W4288797768C177264268 @default.
- W4288797768 hasConceptScore W4288797768C186644900 @default.
- W4288797768 hasConceptScore W4288797768C187736073 @default.
- W4288797768 hasConceptScore W4288797768C199360897 @default.
- W4288797768 hasConceptScore W4288797768C204321447 @default.
- W4288797768 hasConceptScore W4288797768C2776760102 @default.
- W4288797768 hasConceptScore W4288797768C2780451532 @default.
- W4288797768 hasConceptScore W4288797768C2780513914 @default.
- W4288797768 hasConceptScore W4288797768C41008148 @default.
- W4288797768 hasConceptScore W4288797768C75608658 @default.
- W4288797768 hasConceptScore W4288797768C89600930 @default.
- W4288797768 hasLocation W42887977681 @default.
- W4288797768 hasOpenAccess W4288797768 @default.
- W4288797768 hasPrimaryLocation W42887977681 @default.
- W4288797768 hasRelatedWork W1481149166 @default.
- W4288797768 hasRelatedWork W1592748392 @default.
- W4288797768 hasRelatedWork W1940315767 @default.
- W4288797768 hasRelatedWork W2379749588 @default.
- W4288797768 hasRelatedWork W2535890285 @default.
- W4288797768 hasRelatedWork W2885441405 @default.
- W4288797768 hasRelatedWork W2926519392 @default.
- W4288797768 hasRelatedWork W3140547576 @default.
- W4288797768 hasRelatedWork W3202524699 @default.
- W4288797768 hasRelatedWork W4287241433 @default.
- W4288797768 isParatext "false" @default.
- W4288797768 isRetracted "false" @default.
- W4288797768 workType "article" @default.