Matches in SemOpenAlex for { <https://semopenalex.org/work/W3188017601> ?p ?o ?g. }
- W3188017601 abstract "Classical close-set semantic segmentation networks have limited ability to detect out-of-distribution (OOD) objects, which is important for safety-critical applications such as autonomous driving. Incrementally learning these OOD objects with few annotations is an ideal way to enlarge the knowledge base of the deep learning models. In this paper, we propose an open world semantic segmentation system that includes two modules: (1) an open-set semantic segmentation module to detect both in-distribution and OOD objects. (2) an incremental few-shot learning module to gradually incorporate those OOD objects into its existing knowledge base. This open world semantic segmentation system behaves like a human being, which is able to identify OOD objects and gradually learn them with corresponding supervision. We adopt the Deep Metric Learning Network (DMLNet) with contrastive clustering to implement open-set semantic segmentation. Compared to other open-set semantic segmentation methods, our DMLNet achieves state-of-the-art performance on three challenging open-set semantic segmentation datasets without using additional data or generative models. On this basis, two incremental few-shot learning methods are further proposed to progressively improve the DMLNet with the annotations of OOD objects." @default.
- W3188017601 created "2021-08-16" @default.
- W3188017601 creator A5003897229 @default.
- W3188017601 creator A5019137312 @default.
- W3188017601 creator A5068828641 @default.
- W3188017601 creator A5085942352 @default.
- W3188017601 creator A5090316323 @default.
- W3188017601 date "2021-08-10" @default.
- W3188017601 modified "2023-09-27" @default.
- W3188017601 title "Deep Metric Learning for Open World Semantic Segmentation" @default.
- W3188017601 cites W1567041090 @default.
- W3188017601 cites W1682403713 @default.
- W3188017601 cites W2026915078 @default.
- W3188017601 cites W2081462278 @default.
- W3188017601 cites W2135442311 @default.
- W3188017601 cites W2194775991 @default.
- W3188017601 cites W2340897893 @default.
- W3188017601 cites W2531327146 @default.
- W3188017601 cites W2560023338 @default.
- W3188017601 cites W2600383743 @default.
- W3188017601 cites W2601450892 @default.
- W3188017601 cites W2781228439 @default.
- W3188017601 cites W2791763440 @default.
- W3188017601 cites W2796762894 @default.
- W3188017601 cites W2798836702 @default.
- W3188017601 cites W2808859004 @default.
- W3188017601 cites W2910707576 @default.
- W3188017601 cites W2912687289 @default.
- W3188017601 cites W2924485953 @default.
- W3188017601 cites W2962723986 @default.
- W3188017601 cites W2963045681 @default.
- W3188017601 cites W2963238274 @default.
- W3188017601 cites W2963314614 @default.
- W3188017601 cites W2963845150 @default.
- W3188017601 cites W2963944506 @default.
- W3188017601 cites W2964059111 @default.
- W3188017601 cites W2964189064 @default.
- W3188017601 cites W2964309882 @default.
- W3188017601 cites W2981786146 @default.
- W3188017601 cites W2990881247 @default.
- W3188017601 cites W3035513921 @default.
- W3188017601 cites W3035564946 @default.
- W3188017601 cites W3040318838 @default.
- W3188017601 cites W3095099350 @default.
- W3188017601 cites W3102616566 @default.
- W3188017601 cites W3108603282 @default.
- W3188017601 cites W3110437118 @default.
- W3188017601 cites W3113089224 @default.
- W3188017601 cites W3118707261 @default.
- W3188017601 cites W3122923658 @default.
- W3188017601 cites W3176709420 @default.
- W3188017601 cites W4226489086 @default.
- W3188017601 doi "https://doi.org/10.48550/arxiv.2108.04562" @default.
- W3188017601 hasPublicationYear "2021" @default.
- W3188017601 type Work @default.
- W3188017601 sameAs 3188017601 @default.
- W3188017601 citedByCount "1" @default.
- W3188017601 countsByYear W31880176012023 @default.
- W3188017601 crossrefType "posted-content" @default.
- W3188017601 hasAuthorship W3188017601A5003897229 @default.
- W3188017601 hasAuthorship W3188017601A5019137312 @default.
- W3188017601 hasAuthorship W3188017601A5068828641 @default.
- W3188017601 hasAuthorship W3188017601A5085942352 @default.
- W3188017601 hasAuthorship W3188017601A5090316323 @default.
- W3188017601 hasBestOaLocation W31880176011 @default.
- W3188017601 hasConcept C108583219 @default.
- W3188017601 hasConcept C118615104 @default.
- W3188017601 hasConcept C119857082 @default.
- W3188017601 hasConcept C154945302 @default.
- W3188017601 hasConcept C162324750 @default.
- W3188017601 hasConcept C176217482 @default.
- W3188017601 hasConcept C177264268 @default.
- W3188017601 hasConcept C199360897 @default.
- W3188017601 hasConcept C21547014 @default.
- W3188017601 hasConcept C33923547 @default.
- W3188017601 hasConcept C41008148 @default.
- W3188017601 hasConcept C42357961 @default.
- W3188017601 hasConcept C89600930 @default.
- W3188017601 hasConceptScore W3188017601C108583219 @default.
- W3188017601 hasConceptScore W3188017601C118615104 @default.
- W3188017601 hasConceptScore W3188017601C119857082 @default.
- W3188017601 hasConceptScore W3188017601C154945302 @default.
- W3188017601 hasConceptScore W3188017601C162324750 @default.
- W3188017601 hasConceptScore W3188017601C176217482 @default.
- W3188017601 hasConceptScore W3188017601C177264268 @default.
- W3188017601 hasConceptScore W3188017601C199360897 @default.
- W3188017601 hasConceptScore W3188017601C21547014 @default.
- W3188017601 hasConceptScore W3188017601C33923547 @default.
- W3188017601 hasConceptScore W3188017601C41008148 @default.
- W3188017601 hasConceptScore W3188017601C42357961 @default.
- W3188017601 hasConceptScore W3188017601C89600930 @default.
- W3188017601 hasLocation W31880176011 @default.
- W3188017601 hasOpenAccess W3188017601 @default.
- W3188017601 hasPrimaryLocation W31880176011 @default.
- W3188017601 hasRelatedWork W2948658236 @default.
- W3188017601 hasRelatedWork W3014300295 @default.
- W3188017601 hasRelatedWork W4220708658 @default.
- W3188017601 hasRelatedWork W4223943233 @default.
- W3188017601 hasRelatedWork W4225161397 @default.
- W3188017601 hasRelatedWork W4243168368 @default.