Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386075597> ?p ?o ?g. }
- W4386075597 abstract "Existing instance segmentation models learn task-specific information using manual mask annotations from base (training) categories. These mask annotations require tremendous human effort, limiting the scalability to annotate novel (new) categories. To alleviate this problem, Open-Vocabulary (OV) methods leverage large-scale image-caption pairs and vision-language models to learn novel categories. In summary, an OV method learns task-specific information using strong supervision from base annotations and novel category information using weak supervision from image-captions pairs. This difference between strong and weak supervision leads to overfitting on base categories, resulting in poor generalization towards novel categories. In this work, we overcome this issue by learning both base and novel categories from pseudo-mask annotations generated by the vision-language model in a weakly supervised manner using our proposed Mask-free OVIS pipeline. Our method automatically generates pseudo-mask annotations by leveraging the localization ability of a pre-trained vision-language model for objects present in image-caption pairs. The generated pseudo-mask annotations are then used to supervise an instance segmentation model, freeing the entire pipeline from any labour-expensive instance-level annotations and overfitting. Our extensive experiments show that our method trained with just pseudo-masks significantly improves the mAP scores on the MS-COCO dataset and OpenImages dataset compared to the recent state-of-the-art methods trained with manual masks. Codes and models are provided in https://vibashan.github.io/ovis-web/." @default.
- W4386075597 created "2023-08-23" @default.
- W4386075597 creator A5004330920 @default.
- W4386075597 creator A5010420271 @default.
- W4386075597 creator A5018518655 @default.
- W4386075597 creator A5021042598 @default.
- W4386075597 creator A5026664462 @default.
- W4386075597 creator A5033532580 @default.
- W4386075597 creator A5042658750 @default.
- W4386075597 creator A5074905453 @default.
- W4386075597 date "2023-06-01" @default.
- W4386075597 modified "2023-09-27" @default.
- W4386075597 title "Mask-Free OVIS: Open-Vocabulary Instance Segmentation without Manual Mask Annotations" @default.
- W4386075597 cites W2295107390 @default.
- W4386075597 cites W2604113307 @default.
- W4386075597 cites W2940612399 @default.
- W4386075597 cites W2953433552 @default.
- W4386075597 cites W2955278847 @default.
- W4386075597 cites W2962858109 @default.
- W4386075597 cites W2963150697 @default.
- W4386075597 cites W2963311325 @default.
- W4386075597 cites W2963603913 @default.
- W4386075597 cites W2963606198 @default.
- W4386075597 cites W2964236837 @default.
- W4386075597 cites W2964328846 @default.
- W4386075597 cites W2981613027 @default.
- W4386075597 cites W2983061571 @default.
- W4386075597 cites W2991023920 @default.
- W4386075597 cites W2991662170 @default.
- W4386075597 cites W2997998901 @default.
- W4386075597 cites W3034199269 @default.
- W4386075597 cites W3035725370 @default.
- W4386075597 cites W3099193570 @default.
- W4386075597 cites W3119344692 @default.
- W4386075597 cites W3152635971 @default.
- W4386075597 cites W3172507542 @default.
- W4386075597 cites W3173859428 @default.
- W4386075597 cites W3176164117 @default.
- W4386075597 cites W3176692018 @default.
- W4386075597 cites W3180169285 @default.
- W4386075597 cites W3203318343 @default.
- W4386075597 cites W3203360801 @default.
- W4386075597 cites W3216939881 @default.
- W4386075597 cites W4288083516 @default.
- W4386075597 cites W4296899977 @default.
- W4386075597 cites W4312563428 @default.
- W4386075597 cites W4312747482 @default.
- W4386075597 cites W4312890493 @default.
- W4386075597 cites W4312956471 @default.
- W4386075597 doi "https://doi.org/10.1109/cvpr52729.2023.02254" @default.
- W4386075597 hasPublicationYear "2023" @default.
- W4386075597 type Work @default.
- W4386075597 citedByCount "0" @default.
- W4386075597 crossrefType "proceedings-article" @default.
- W4386075597 hasAuthorship W4386075597A5004330920 @default.
- W4386075597 hasAuthorship W4386075597A5010420271 @default.
- W4386075597 hasAuthorship W4386075597A5018518655 @default.
- W4386075597 hasAuthorship W4386075597A5021042598 @default.
- W4386075597 hasAuthorship W4386075597A5026664462 @default.
- W4386075597 hasAuthorship W4386075597A5033532580 @default.
- W4386075597 hasAuthorship W4386075597A5042658750 @default.
- W4386075597 hasAuthorship W4386075597A5074905453 @default.
- W4386075597 hasConcept C119857082 @default.
- W4386075597 hasConcept C138885662 @default.
- W4386075597 hasConcept C153083717 @default.
- W4386075597 hasConcept C153180895 @default.
- W4386075597 hasConcept C154945302 @default.
- W4386075597 hasConcept C162324750 @default.
- W4386075597 hasConcept C187736073 @default.
- W4386075597 hasConcept C199360897 @default.
- W4386075597 hasConcept C204321447 @default.
- W4386075597 hasConcept C22019652 @default.
- W4386075597 hasConcept C2776321320 @default.
- W4386075597 hasConcept C2777601683 @default.
- W4386075597 hasConcept C2780451532 @default.
- W4386075597 hasConcept C31972630 @default.
- W4386075597 hasConcept C41008148 @default.
- W4386075597 hasConcept C41895202 @default.
- W4386075597 hasConcept C43521106 @default.
- W4386075597 hasConcept C48044578 @default.
- W4386075597 hasConcept C50644808 @default.
- W4386075597 hasConcept C77088390 @default.
- W4386075597 hasConcept C89600930 @default.
- W4386075597 hasConceptScore W4386075597C119857082 @default.
- W4386075597 hasConceptScore W4386075597C138885662 @default.
- W4386075597 hasConceptScore W4386075597C153083717 @default.
- W4386075597 hasConceptScore W4386075597C153180895 @default.
- W4386075597 hasConceptScore W4386075597C154945302 @default.
- W4386075597 hasConceptScore W4386075597C162324750 @default.
- W4386075597 hasConceptScore W4386075597C187736073 @default.
- W4386075597 hasConceptScore W4386075597C199360897 @default.
- W4386075597 hasConceptScore W4386075597C204321447 @default.
- W4386075597 hasConceptScore W4386075597C22019652 @default.
- W4386075597 hasConceptScore W4386075597C2776321320 @default.
- W4386075597 hasConceptScore W4386075597C2777601683 @default.
- W4386075597 hasConceptScore W4386075597C2780451532 @default.
- W4386075597 hasConceptScore W4386075597C31972630 @default.
- W4386075597 hasConceptScore W4386075597C41008148 @default.
- W4386075597 hasConceptScore W4386075597C41895202 @default.
- W4386075597 hasConceptScore W4386075597C43521106 @default.