Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386076130> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W4386076130 abstract "Humans excel at learning from expert demonstrations and solving their own problems. To equip intelligent robots and assistants, such as AR glasses, with this ability, it is essential to ground human hand interactions (i.e., affordances) from demonstration videos and apply them to a target image like a user's AR glass view. This video-to-image affordance grounding task is challenging due to (1) the need to predict fine-grained affordances, and (2) the limited training data, which inadequately covers video-image discrepancies and negatively impacts grounding. To tackle them, we propose Affordance Transformer (Afformer), which has a fine-grained transformer-based decoder that gradually refines affordance grounding. Moreover, we introduce Mask Affordance Hand (MaskAHand), a self-supervised pre-training technique for synthesizing video-image data and simulating context changes, enhancing affordance grounding across video-image discrepancies. Afformer with MaskAHand pre-training achieves state-of-the-art performance on multiple benchmarks, including a sub-stantial 37% improvement on the OPRA dataset. Code is made available at https://github.com/showlab/afformer." @default.
- W4386076130 created "2023-08-23" @default.
- W4386076130 creator A5000312791 @default.
- W4386076130 creator A5001133932 @default.
- W4386076130 creator A5068937750 @default.
- W4386076130 creator A5083481253 @default.
- W4386076130 date "2023-06-01" @default.
- W4386076130 modified "2023-09-27" @default.
- W4386076130 title "Affordance Grounding from Demonstration Video to Target Image" @default.
- W4386076130 cites W1522734439 @default.
- W4386076130 cites W2055452679 @default.
- W4386076130 cites W2151103935 @default.
- W4386076130 cites W2194775991 @default.
- W4386076130 cites W2565639579 @default.
- W4386076130 cites W2763331099 @default.
- W4386076130 cites W2798897443 @default.
- W4386076130 cites W2799067027 @default.
- W4386076130 cites W2888096830 @default.
- W4386076130 cites W2962858109 @default.
- W4386076130 cites W2963503775 @default.
- W4386076130 cites W2963828885 @default.
- W4386076130 cites W2983465317 @default.
- W4386076130 cites W2990503944 @default.
- W4386076130 cites W3034891989 @default.
- W4386076130 cites W3043075211 @default.
- W4386076130 cites W3084413476 @default.
- W4386076130 cites W3131500599 @default.
- W4386076130 cites W3175450634 @default.
- W4386076130 cites W3207908209 @default.
- W4386076130 cites W3217102353 @default.
- W4386076130 cites W4221153063 @default.
- W4386076130 cites W4290546024 @default.
- W4386076130 cites W4312312750 @default.
- W4386076130 cites W4312769131 @default.
- W4386076130 cites W4312769771 @default.
- W4386076130 cites W4312815172 @default.
- W4386076130 cites W4312903454 @default.
- W4386076130 cites W4312930076 @default.
- W4386076130 cites W4313156423 @default.
- W4386076130 cites W4385573410 @default.
- W4386076130 cites W639708223 @default.
- W4386076130 doi "https://doi.org/10.1109/cvpr52729.2023.00657" @default.
- W4386076130 hasPublicationYear "2023" @default.
- W4386076130 type Work @default.
- W4386076130 citedByCount "0" @default.
- W4386076130 crossrefType "proceedings-article" @default.
- W4386076130 hasAuthorship W4386076130A5000312791 @default.
- W4386076130 hasAuthorship W4386076130A5001133932 @default.
- W4386076130 hasAuthorship W4386076130A5068937750 @default.
- W4386076130 hasAuthorship W4386076130A5083481253 @default.
- W4386076130 hasConcept C107457646 @default.
- W4386076130 hasConcept C119599485 @default.
- W4386076130 hasConcept C127413603 @default.
- W4386076130 hasConcept C151730666 @default.
- W4386076130 hasConcept C154945302 @default.
- W4386076130 hasConcept C165801399 @default.
- W4386076130 hasConcept C168993435 @default.
- W4386076130 hasConcept C194995250 @default.
- W4386076130 hasConcept C201995342 @default.
- W4386076130 hasConcept C2779343474 @default.
- W4386076130 hasConcept C2780451532 @default.
- W4386076130 hasConcept C31972630 @default.
- W4386076130 hasConcept C41008148 @default.
- W4386076130 hasConcept C66322947 @default.
- W4386076130 hasConcept C86803240 @default.
- W4386076130 hasConceptScore W4386076130C107457646 @default.
- W4386076130 hasConceptScore W4386076130C119599485 @default.
- W4386076130 hasConceptScore W4386076130C127413603 @default.
- W4386076130 hasConceptScore W4386076130C151730666 @default.
- W4386076130 hasConceptScore W4386076130C154945302 @default.
- W4386076130 hasConceptScore W4386076130C165801399 @default.
- W4386076130 hasConceptScore W4386076130C168993435 @default.
- W4386076130 hasConceptScore W4386076130C194995250 @default.
- W4386076130 hasConceptScore W4386076130C201995342 @default.
- W4386076130 hasConceptScore W4386076130C2779343474 @default.
- W4386076130 hasConceptScore W4386076130C2780451532 @default.
- W4386076130 hasConceptScore W4386076130C31972630 @default.
- W4386076130 hasConceptScore W4386076130C41008148 @default.
- W4386076130 hasConceptScore W4386076130C66322947 @default.
- W4386076130 hasConceptScore W4386076130C86803240 @default.
- W4386076130 hasFunder F4320320709 @default.
- W4386076130 hasLocation W43860761301 @default.
- W4386076130 hasOpenAccess W4386076130 @default.
- W4386076130 hasPrimaryLocation W43860761301 @default.
- W4386076130 hasRelatedWork W1504101963 @default.
- W4386076130 hasRelatedWork W1559800691 @default.
- W4386076130 hasRelatedWork W1891287906 @default.
- W4386076130 hasRelatedWork W1969923398 @default.
- W4386076130 hasRelatedWork W2036807459 @default.
- W4386076130 hasRelatedWork W2166024367 @default.
- W4386076130 hasRelatedWork W2417026147 @default.
- W4386076130 hasRelatedWork W2755342338 @default.
- W4386076130 hasRelatedWork W2772917594 @default.
- W4386076130 hasRelatedWork W2775347418 @default.
- W4386076130 isParatext "false" @default.
- W4386076130 isRetracted "false" @default.
- W4386076130 workType "article" @default.