Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312748662> ?p ?o ?g. }
Showing items 1 to 100 of
100
with 100 items per page.
- W4312748662 endingPage "724" @default.
- W4312748662 startingPage "708" @default.
- W4312748662 abstract "Weakly-supervised object detection (WSOD) aims to train an object detector only requiring the image-level annotations. Recently, some works have managed to select the accurate boxes generated from a well-trained WSOD network to supervise a semi-supervised detection framework for better performance. However, these approaches simply divide the training set into labeled and unlabeled sets according to the image-level criteria, such that sufficient mislabeled or wrongly localized box predictions are chosen as pseudo ground-truths, resulting in a sub-optimal solution of detection performance. To overcome this issue, we propose a novel WSOD framework with a new paradigm that switches from weak supervision to noisy supervision (W2N). Generally, with given pseudo ground-truths generated from the well-trained WSOD network, we propose a two-module iterative training algorithm to refine pseudo labels and supervise better object detector progressively. In the localization adaptation module, we propose a regularization loss to reduce the proportion of discriminative parts in original pseudo ground-truths, obtaining better pseudo ground-truths for further training. In the semi-supervised module, we propose a two tasks instance-level split method to select high-quality labels for training a semi-supervised detector. Experimental results on different benchmarks verify the effectiveness of W2N, and our W2N outperforms all existing pure WSOD methods and transfer learning methods. Our code is publicly available at https://github.com/1170300714/w2n_wsod ." @default.
- W4312748662 created "2023-01-05" @default.
- W4312748662 creator A5018318136 @default.
- W4312748662 creator A5060050139 @default.
- W4312748662 creator A5066588572 @default.
- W4312748662 creator A5075762321 @default.
- W4312748662 creator A5078813046 @default.
- W4312748662 date "2022-01-01" @default.
- W4312748662 modified "2023-10-10" @default.
- W4312748662 title "W2N: Switching from Weak Supervision to Noisy Supervision for Object Detection" @default.
- W4312748662 cites W1536680647 @default.
- W4312748662 cites W1861492603 @default.
- W4312748662 cites W1945443124 @default.
- W4312748662 cites W2031489346 @default.
- W4312748662 cites W2088049833 @default.
- W4312748662 cites W2108598243 @default.
- W4312748662 cites W2110119381 @default.
- W4312748662 cites W2133324800 @default.
- W4312748662 cites W2441255125 @default.
- W4312748662 cites W2604260814 @default.
- W4312748662 cites W2798269247 @default.
- W4312748662 cites W2798748179 @default.
- W4312748662 cites W2813911573 @default.
- W4312748662 cites W2940046420 @default.
- W4312748662 cites W2954087924 @default.
- W4312748662 cites W2962685835 @default.
- W4312748662 cites W2963603913 @default.
- W4312748662 cites W2963943295 @default.
- W4312748662 cites W2963952323 @default.
- W4312748662 cites W2964328846 @default.
- W4312748662 cites W2991023920 @default.
- W4312748662 cites W2991533779 @default.
- W4312748662 cites W2991662170 @default.
- W4312748662 cites W2994041372 @default.
- W4312748662 cites W3034329658 @default.
- W4312748662 cites W3035725370 @default.
- W4312748662 cites W3109576564 @default.
- W4312748662 cites W3172507542 @default.
- W4312748662 cites W3175385356 @default.
- W4312748662 cites W3175704361 @default.
- W4312748662 cites W3190805089 @default.
- W4312748662 cites W3194159212 @default.
- W4312748662 cites W4312866875 @default.
- W4312748662 doi "https://doi.org/10.1007/978-3-031-20056-4_41" @default.
- W4312748662 hasPublicationYear "2022" @default.
- W4312748662 type Work @default.
- W4312748662 citedByCount "3" @default.
- W4312748662 countsByYear W43127486622023 @default.
- W4312748662 crossrefType "book-chapter" @default.
- W4312748662 hasAuthorship W4312748662A5018318136 @default.
- W4312748662 hasAuthorship W4312748662A5060050139 @default.
- W4312748662 hasAuthorship W4312748662A5066588572 @default.
- W4312748662 hasAuthorship W4312748662A5075762321 @default.
- W4312748662 hasAuthorship W4312748662A5078813046 @default.
- W4312748662 hasBestOaLocation W43127486622 @default.
- W4312748662 hasConcept C119857082 @default.
- W4312748662 hasConcept C153180895 @default.
- W4312748662 hasConcept C154945302 @default.
- W4312748662 hasConcept C177264268 @default.
- W4312748662 hasConcept C199360897 @default.
- W4312748662 hasConcept C2776135515 @default.
- W4312748662 hasConcept C2776151529 @default.
- W4312748662 hasConcept C2776760102 @default.
- W4312748662 hasConcept C2781238097 @default.
- W4312748662 hasConcept C41008148 @default.
- W4312748662 hasConcept C76155785 @default.
- W4312748662 hasConcept C94915269 @default.
- W4312748662 hasConcept C97931131 @default.
- W4312748662 hasConceptScore W4312748662C119857082 @default.
- W4312748662 hasConceptScore W4312748662C153180895 @default.
- W4312748662 hasConceptScore W4312748662C154945302 @default.
- W4312748662 hasConceptScore W4312748662C177264268 @default.
- W4312748662 hasConceptScore W4312748662C199360897 @default.
- W4312748662 hasConceptScore W4312748662C2776135515 @default.
- W4312748662 hasConceptScore W4312748662C2776151529 @default.
- W4312748662 hasConceptScore W4312748662C2776760102 @default.
- W4312748662 hasConceptScore W4312748662C2781238097 @default.
- W4312748662 hasConceptScore W4312748662C41008148 @default.
- W4312748662 hasConceptScore W4312748662C76155785 @default.
- W4312748662 hasConceptScore W4312748662C94915269 @default.
- W4312748662 hasConceptScore W4312748662C97931131 @default.
- W4312748662 hasLocation W43127486621 @default.
- W4312748662 hasLocation W43127486622 @default.
- W4312748662 hasOpenAccess W4312748662 @default.
- W4312748662 hasPrimaryLocation W43127486621 @default.
- W4312748662 hasRelatedWork W1972656095 @default.
- W4312748662 hasRelatedWork W2024160000 @default.
- W4312748662 hasRelatedWork W2061273563 @default.
- W4312748662 hasRelatedWork W2285052147 @default.
- W4312748662 hasRelatedWork W2729514902 @default.
- W4312748662 hasRelatedWork W2743258233 @default.
- W4312748662 hasRelatedWork W2773500201 @default.
- W4312748662 hasRelatedWork W2970216048 @default.
- W4312748662 hasRelatedWork W2998168123 @default.
- W4312748662 hasRelatedWork W4287995534 @default.
- W4312748662 isParatext "false" @default.
- W4312748662 isRetracted "false" @default.
- W4312748662 workType "book-chapter" @default.