FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3212849462> ?p ?o ?g. }

• W3212849462 abstract "We study the problem of learning to estimate the 3D object pose from a few labelled examples and a collection of unlabelled data. Our main contribution is a learning framework, neural view synthesis and matching, that can transfer the 3D pose annotation from the labelled to unlabelled images reliably, despite unseen 3D views and nuisance variations such as the object shape, texture, illumination or scene context. In our approach, objects are represented as 3D cuboid meshes composed of feature vectors at each mesh vertex. The model is initialized from a few labelled images and is subsequently used to synthesize feature representations of unseen 3D views. The synthesized views are matched with the feature representations of unlabelled images to generate pseudo-labels of the 3D pose. The pseudo-labelled data is, in turn, used to train the feature extractor such that the features at each mesh vertex are more invariant across varying 3D views of the object. Our model is trained in an EM-type manner alternating between increasing the 3D pose invariance of the feature extractor and annotating unlabelled data through neural view synthesis and matching. We demonstrate the effectiveness of the proposed semi-supervised learning framework for 3D pose estimation on the PASCAL3D+ and KITTI datasets. We find that our approach outperforms all baselines by a wide margin, particularly in an extreme few-shot setting where only 7 annotated images are given. Remarkably, we observe that our model also achieves an exceptional robustness in out-of-distribution scenarios that involve partial occlusion." @default.
• W3212849462 created "2021-11-22" @default.
• W3212849462 creator A5030206677 @default.
• W3212849462 creator A5081606526 @default.
• W3212849462 creator A5084928022 @default.
• W3212849462 creator A5086706224 @default.
• W3212849462 date "2021-10-27" @default.
• W3212849462 modified "2023-09-26" @default.
• W3212849462 title "Neural View Synthesis and Matching for Semi-Supervised Few-Shot Learning of 3D Pose" @default.
• W3212849462 cites W1869500417 @default.
• W3212849462 cites W1949483711 @default.
• W3212849462 cites W1986534176 @default.
• W3212849462 cites W1991264156 @default.
• W3212849462 cites W1991544872 @default.
• W3212849462 cites W2062118960 @default.
• W3212849462 cites W2079057609 @default.
• W3212849462 cites W2099046646 @default.
• W3212849462 cites W2106401878 @default.
• W3212849462 cites W2108598243 @default.
• W3212849462 cites W2111308925 @default.
• W3212849462 cites W2115579991 @default.
• W3212849462 cites W2136504847 @default.
• W3212849462 cites W2145023731 @default.
• W3212849462 cites W2150066425 @default.
• W3212849462 cites W2151103935 @default.
• W3212849462 cites W2157656099 @default.
• W3212849462 cites W2158305599 @default.
• W3212849462 cites W2194775991 @default.
• W3212849462 cites W2203450678 @default.
• W3212849462 cites W2237250383 @default.
• W3212849462 cites W2519887557 @default.
• W3212849462 cites W2560544142 @default.
• W3212849462 cites W2598199894 @default.
• W3212849462 cites W2600447016 @default.
• W3212849462 cites W2612112834 @default.
• W3212849462 cites W2795096917 @default.
• W3212849462 cites W2963325280 @default.
• W3212849462 cites W2963680240 @default.
• W3212849462 cites W2963713828 @default.
• W3212849462 cites W2963749936 @default.
• W3212849462 cites W2963826402 @default.
• W3212849462 cites W2963892972 @default.
• W3212849462 cites W2963996492 @default.
• W3212849462 cites W2968993450 @default.
• W3212849462 cites W2970971581 @default.
• W3212849462 cites W2984353870 @default.
• W3212849462 cites W2996670291 @default.
• W3212849462 cites W3001197829 @default.
• W3212849462 cites W3009561768 @default.
• W3212849462 cites W3010559878 @default.
• W3212849462 cites W3011334834 @default.
• W3212849462 cites W3034973071 @default.
• W3212849462 cites W3035069238 @default.
• W3212849462 cites W3042719542 @default.
• W3212849462 cites W3091655598 @default.
• W3212849462 cites W3107828565 @default.
• W3212849462 cites W3112108866 @default.
• W3212849462 cites W3122672262 @default.
• W3212849462 cites W3132885850 @default.
• W3212849462 cites W3146426667 @default.
• W3212849462 cites W3159772287 @default.
• W3212849462 cites W3172316935 @default.
• W3212849462 cites W3141898517 @default.
• W3212849462 doi "https://doi.org/10.48550/arxiv.2110.14213" @default.
• W3212849462 hasPublicationYear "2021" @default.
• W3212849462 type Work @default.
• W3212849462 sameAs 3212849462 @default.
• W3212849462 citedByCount "0" @default.
• W3212849462 crossrefType "posted-content" @default.
• W3212849462 hasAuthorship W3212849462A5030206677 @default.
• W3212849462 hasAuthorship W3212849462A5081606526 @default.
• W3212849462 hasAuthorship W3212849462A5084928022 @default.
• W3212849462 hasAuthorship W3212849462A5086706224 @default.
• W3212849462 hasBestOaLocation W32128494621 @default.
• W3212849462 hasConcept C104317684 @default.
• W3212849462 hasConcept C105795698 @default.
• W3212849462 hasConcept C119857082 @default.
• W3212849462 hasConcept C121684516 @default.
• W3212849462 hasConcept C138885662 @default.
• W3212849462 hasConcept C153180895 @default.
• W3212849462 hasConcept C154945302 @default.
• W3212849462 hasConcept C165064840 @default.
• W3212849462 hasConcept C185592680 @default.
• W3212849462 hasConcept C2776401178 @default.
• W3212849462 hasConcept C31487907 @default.
• W3212849462 hasConcept C31972630 @default.
• W3212849462 hasConcept C33923547 @default.
• W3212849462 hasConcept C41008148 @default.
• W3212849462 hasConcept C41895202 @default.
• W3212849462 hasConcept C52102323 @default.
• W3212849462 hasConcept C55493867 @default.
• W3212849462 hasConcept C63479239 @default.
• W3212849462 hasConcept C774472 @default.
• W3212849462 hasConceptScore W3212849462C104317684 @default.
• W3212849462 hasConceptScore W3212849462C105795698 @default.
• W3212849462 hasConceptScore W3212849462C119857082 @default.
• W3212849462 hasConceptScore W3212849462C121684516 @default.
• W3212849462 hasConceptScore W3212849462C138885662 @default.
• W3212849462 hasConceptScore W3212849462C153180895 @default.
• W3212849462 hasConceptScore W3212849462C154945302 @default.

• next