Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386075819> ?p ?o ?g. }
- W4386075819 abstract "We present ODISE: Open-vocabulary DIffusion-based panoptic SEgmentation, which unifies pre-trained text-image diffusion and discriminative models to perform open-vocabulary panoptic segmentation. Text-to-image diffusion models have the remarkable ability to generate high-quality images with diverse open-vocabulary language descriptions. This demonstrates that their internal representation space is highly correlated with open concepts in the real world. Text-image discriminative models like CLIP, on the other hand, are good at classifying images into open-vocabulary labels. We leverage the frozen internal representations of both these models to perform panoptic segmentation of any category in the wild. Our approach outperforms the previous state of the art by significant margins on both open-vocabulary panoptic and semantic segmentation tasks. In particular, with COCO training only, our method achieves 23.4 PQ and 30.0 mIoU on the ADE20K dataset, with 8.3 PQ and 7.9 mIoU absolute improvement over the previous state of the art. We open-source our code and models at https://github.com/NVlabs/ODISE." @default.
- W4386075819 created "2023-08-23" @default.
- W4386075819 creator A5011351573 @default.
- W4386075819 creator A5021623206 @default.
- W4386075819 creator A5025693806 @default.
- W4386075819 creator A5038984764 @default.
- W4386075819 creator A5049815485 @default.
- W4386075819 creator A5066028215 @default.
- W4386075819 date "2023-06-01" @default.
- W4386075819 modified "2023-10-16" @default.
- W4386075819 title "Open-Vocabulary Panoptic Segmentation with Text-to-Image Diffusion Models" @default.
- W4386075819 cites W2031489346 @default.
- W4386075819 cites W2108598243 @default.
- W4386075819 cites W2125215748 @default.
- W4386075819 cites W2194775991 @default.
- W4386075819 cites W2507296351 @default.
- W4386075819 cites W2565639579 @default.
- W4386075819 cites W2910628332 @default.
- W4386075819 cites W2924485953 @default.
- W4386075819 cites W2962770929 @default.
- W4386075819 cites W2962793481 @default.
- W4386075819 cites W2963150697 @default.
- W4386075819 cites W2999219213 @default.
- W4386075819 cites W3034355852 @default.
- W4386075819 cites W3035574324 @default.
- W4386075819 cites W3138516171 @default.
- W4386075819 cites W3145450063 @default.
- W4386075819 cites W3159116077 @default.
- W4386075819 cites W3159481202 @default.
- W4386075819 cites W3167788848 @default.
- W4386075819 cites W3168649818 @default.
- W4386075819 cites W3170410843 @default.
- W4386075819 cites W3173143063 @default.
- W4386075819 cites W3173397696 @default.
- W4386075819 cites W3173859428 @default.
- W4386075819 cites W3176659256 @default.
- W4386075819 cites W3180355996 @default.
- W4386075819 cites W4286982960 @default.
- W4386075819 cites W4312438583 @default.
- W4386075819 cites W4312443924 @default.
- W4386075819 cites W4312563428 @default.
- W4386075819 cites W4312747482 @default.
- W4386075819 cites W4312769131 @default.
- W4386075819 cites W4312815172 @default.
- W4386075819 cites W4312933868 @default.
- W4386075819 cites W4312980231 @default.
- W4386075819 cites W4313156423 @default.
- W4386075819 doi "https://doi.org/10.1109/cvpr52729.2023.00289" @default.
- W4386075819 hasPublicationYear "2023" @default.
- W4386075819 type Work @default.
- W4386075819 citedByCount "0" @default.
- W4386075819 crossrefType "proceedings-article" @default.
- W4386075819 hasAuthorship W4386075819A5011351573 @default.
- W4386075819 hasAuthorship W4386075819A5021623206 @default.
- W4386075819 hasAuthorship W4386075819A5025693806 @default.
- W4386075819 hasAuthorship W4386075819A5038984764 @default.
- W4386075819 hasAuthorship W4386075819A5049815485 @default.
- W4386075819 hasAuthorship W4386075819A5066028215 @default.
- W4386075819 hasConcept C124504099 @default.
- W4386075819 hasConcept C138569888 @default.
- W4386075819 hasConcept C138885662 @default.
- W4386075819 hasConcept C153180895 @default.
- W4386075819 hasConcept C154945302 @default.
- W4386075819 hasConcept C17744445 @default.
- W4386075819 hasConcept C199539241 @default.
- W4386075819 hasConcept C204321447 @default.
- W4386075819 hasConcept C2777601683 @default.
- W4386075819 hasConcept C31972630 @default.
- W4386075819 hasConcept C41008148 @default.
- W4386075819 hasConcept C41895202 @default.
- W4386075819 hasConcept C89600930 @default.
- W4386075819 hasConcept C94625758 @default.
- W4386075819 hasConcept C97931131 @default.
- W4386075819 hasConceptScore W4386075819C124504099 @default.
- W4386075819 hasConceptScore W4386075819C138569888 @default.
- W4386075819 hasConceptScore W4386075819C138885662 @default.
- W4386075819 hasConceptScore W4386075819C153180895 @default.
- W4386075819 hasConceptScore W4386075819C154945302 @default.
- W4386075819 hasConceptScore W4386075819C17744445 @default.
- W4386075819 hasConceptScore W4386075819C199539241 @default.
- W4386075819 hasConceptScore W4386075819C204321447 @default.
- W4386075819 hasConceptScore W4386075819C2777601683 @default.
- W4386075819 hasConceptScore W4386075819C31972630 @default.
- W4386075819 hasConceptScore W4386075819C41008148 @default.
- W4386075819 hasConceptScore W4386075819C41895202 @default.
- W4386075819 hasConceptScore W4386075819C89600930 @default.
- W4386075819 hasConceptScore W4386075819C94625758 @default.
- W4386075819 hasConceptScore W4386075819C97931131 @default.
- W4386075819 hasLocation W43860758191 @default.
- W4386075819 hasOpenAccess W4386075819 @default.
- W4386075819 hasPrimaryLocation W43860758191 @default.
- W4386075819 hasRelatedWork W1669643531 @default.
- W4386075819 hasRelatedWork W2005437358 @default.
- W4386075819 hasRelatedWork W2024160000 @default.
- W4386075819 hasRelatedWork W2061273563 @default.
- W4386075819 hasRelatedWork W2285052147 @default.
- W4386075819 hasRelatedWork W2517104666 @default.
- W4386075819 hasRelatedWork W2729514902 @default.
- W4386075819 hasRelatedWork W2773500201 @default.
- W4386075819 hasRelatedWork W4287995534 @default.