FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3197083248> ?p ?o ?g. }

• W3197083248 abstract "Autonomous driving has attracted much attention over the years but turns out to be harder than expected, probably due to the difficulty of labeled data collection for model training. Self-supervised learning (SSL), which leverages unlabeled data only for representation learning, might be a promising way to improve model performance. Existing SSL methods, however, usually rely on the single-centric-object guarantee, which may not be applicable for multi-instance datasets such as street scenes. To alleviate this limitation, we raise two issues to solve: (1) how to define positive samples for cross-view consistency and (2) how to measure similarity in multi-instance circumstances. We first adopt an IoU threshold during random cropping to transfer global-inconsistency to local-consistency. Then, we propose two feature alignment methods to enable 2D feature maps for multi-instance similarity measurement. Additionally, we adopt intra-image clustering with self-attention for further mining intra-image similarity and translation-invariance. Experiments show that, when pre-trained on Waymo dataset, our method called Multi-instance Siamese Network (MultiSiam) remarkably improves generalization ability and achieves state-of-the-art transfer performance on autonomous driving benchmarks, including Cityscapes and BDD100K, while existing SSL counterparts like MoCo, MoCo-v2, and BYOL show significant performance drop. By pre-training on SODA10M, a large-scale autonomous driving dataset, MultiSiam exceeds the ImageNet pre-trained MoCo-v2, demonstrating the potential of domain-specific pre-training. Code will be available at this https URL." @default.
• W3197083248 created "2021-09-13" @default.
• W3197083248 creator A5003882618 @default.
• W3197083248 creator A5041457457 @default.
• W3197083248 creator A5073139380 @default.
• W3197083248 creator A5074339191 @default.
• W3197083248 creator A5077862962 @default.
• W3197083248 date "2021-08-27" @default.
• W3197083248 modified "2023-09-25" @default.
• W3197083248 title "MultiSiam: Self-supervised Multi-instance Siamese Representation Learning for Autonomous Driving" @default.
• W3197083248 cites W1861492603 @default.
• W3197083248 cites W1903029394 @default.
• W3197083248 cites W2031489346 @default.
• W3197083248 cites W2108598243 @default.
• W3197083248 cites W2194775991 @default.
• W3197083248 cites W2340897893 @default.
• W3197083248 cites W2588289902 @default.
• W3197083248 cites W2757910899 @default.
• W3197083248 cites W2798991696 @default.
• W3197083248 cites W2799352588 @default.
• W3197083248 cites W2842511635 @default.
• W3197083248 cites W2883725317 @default.
• W3197083248 cites W2949517790 @default.
• W3197083248 cites W2953303875 @default.
• W3197083248 cites W2963091558 @default.
• W3197083248 cites W2963623257 @default.
• W3197083248 cites W2987741655 @default.
• W3197083248 cites W2995181141 @default.
• W3197083248 cites W3005680577 @default.
• W3197083248 cites W3009561768 @default.
• W3197083248 cites W3034781633 @default.
• W3197083248 cites W3035060554 @default.
• W3197083248 cites W3035172746 @default.
• W3197083248 cites W3035524453 @default.
• W3197083248 cites W3036224891 @default.
• W3197083248 cites W3040318838 @default.
• W3197083248 cites W3099634177 @default.
• W3197083248 cites W3106428938 @default.
• W3197083248 cites W3106528393 @default.
• W3197083248 cites W3107668149 @default.
• W3197083248 cites W3128661784 @default.
• W3197083248 cites W3173895644 @default.
• W3197083248 hasPublicationYear "2021" @default.
• W3197083248 type Work @default.
• W3197083248 sameAs 3197083248 @default.
• W3197083248 citedByCount "0" @default.
• W3197083248 crossrefType "posted-content" @default.
• W3197083248 hasAuthorship W3197083248A5003882618 @default.
• W3197083248 hasAuthorship W3197083248A5041457457 @default.
• W3197083248 hasAuthorship W3197083248A5073139380 @default.
• W3197083248 hasAuthorship W3197083248A5074339191 @default.
• W3197083248 hasAuthorship W3197083248A5077862962 @default.
• W3197083248 hasConcept C103278499 @default.
• W3197083248 hasConcept C115961682 @default.
• W3197083248 hasConcept C119857082 @default.
• W3197083248 hasConcept C124101348 @default.
• W3197083248 hasConcept C134306372 @default.
• W3197083248 hasConcept C138885662 @default.
• W3197083248 hasConcept C150899416 @default.
• W3197083248 hasConcept C153180895 @default.
• W3197083248 hasConcept C154945302 @default.
• W3197083248 hasConcept C177148314 @default.
• W3197083248 hasConcept C177264268 @default.
• W3197083248 hasConcept C17744445 @default.
• W3197083248 hasConcept C199360897 @default.
• W3197083248 hasConcept C199539241 @default.
• W3197083248 hasConcept C2776359362 @default.
• W3197083248 hasConcept C2776401178 @default.
• W3197083248 hasConcept C2776436953 @default.
• W3197083248 hasConcept C2776760102 @default.
• W3197083248 hasConcept C33923547 @default.
• W3197083248 hasConcept C41008148 @default.
• W3197083248 hasConcept C41895202 @default.
• W3197083248 hasConcept C59404180 @default.
• W3197083248 hasConcept C73555534 @default.
• W3197083248 hasConcept C94625758 @default.
• W3197083248 hasConceptScore W3197083248C103278499 @default.
• W3197083248 hasConceptScore W3197083248C115961682 @default.
• W3197083248 hasConceptScore W3197083248C119857082 @default.
• W3197083248 hasConceptScore W3197083248C124101348 @default.
• W3197083248 hasConceptScore W3197083248C134306372 @default.
• W3197083248 hasConceptScore W3197083248C138885662 @default.
• W3197083248 hasConceptScore W3197083248C150899416 @default.
• W3197083248 hasConceptScore W3197083248C153180895 @default.
• W3197083248 hasConceptScore W3197083248C154945302 @default.
• W3197083248 hasConceptScore W3197083248C177148314 @default.
• W3197083248 hasConceptScore W3197083248C177264268 @default.
• W3197083248 hasConceptScore W3197083248C17744445 @default.
• W3197083248 hasConceptScore W3197083248C199360897 @default.
• W3197083248 hasConceptScore W3197083248C199539241 @default.
• W3197083248 hasConceptScore W3197083248C2776359362 @default.
• W3197083248 hasConceptScore W3197083248C2776401178 @default.
• W3197083248 hasConceptScore W3197083248C2776436953 @default.
• W3197083248 hasConceptScore W3197083248C2776760102 @default.
• W3197083248 hasConceptScore W3197083248C33923547 @default.
• W3197083248 hasConceptScore W3197083248C41008148 @default.
• W3197083248 hasConceptScore W3197083248C41895202 @default.
• W3197083248 hasConceptScore W3197083248C59404180 @default.
• W3197083248 hasConceptScore W3197083248C73555534 @default.
• W3197083248 hasConceptScore W3197083248C94625758 @default.

• next