Matches in SemOpenAlex for { <https://semopenalex.org/work/W2951617797> ?p ?o ?g. }
- W2951617797 abstract "Manually labeling datasets with object masks is extremely time consuming. In this work, we follow the idea of Polygon-RNN to produce polygonal annotations of objects interactively using humans-in-the-loop. We introduce several important improvements to the model: 1) we design a new CNN encoder architecture, 2) show how to effectively train the model with Reinforcement Learning, and 3) significantly increase the output resolution using a Graph Neural Network, allowing the model to accurately annotate high-resolution objects in images. Extensive evaluation on the Cityscapes dataset shows that our model, which we refer to as Polygon-RNN++, significantly outperforms the original model in both automatic (10% absolute and 16% relative improvement in mean IoU) and interactive modes (requiring 50% fewer clicks by annotators). We further analyze the cross-domain scenario in which our model is trained on one dataset, and used out of the box on datasets from varying domains. The results show that Polygon-RNN++ exhibits powerful generalization capabilities, achieving significant improvements over existing pixel-wise methods. Using simple online fine-tuning we further achieve a high reduction in annotation time for new datasets, moving a step closer towards an interactive annotation tool to be used in practice." @default.
- W2951617797 created "2019-06-27" @default.
- W2951617797 creator A5030331254 @default.
- W2951617797 creator A5043244870 @default.
- W2951617797 creator A5070642269 @default.
- W2951617797 creator A5078831654 @default.
- W2951617797 date "2018-03-26" @default.
- W2951617797 modified "2023-09-29" @default.
- W2951617797 title "Efficient Interactive Annotation of Segmentation Datasets with Polygon-RNN++" @default.
- W2951617797 cites W1756252575 @default.
- W2951617797 cites W1903029394 @default.
- W2951617797 cites W2009921846 @default.
- W2951617797 cites W2059319345 @default.
- W2951617797 cites W2065768018 @default.
- W2951617797 cites W2113137767 @default.
- W2951617797 cites W2116341502 @default.
- W2951617797 cites W2117775864 @default.
- W2951617797 cites W2119717200 @default.
- W2951617797 cites W2124351162 @default.
- W2951617797 cites W2150066425 @default.
- W2951617797 cites W2169551590 @default.
- W2951617797 cites W2175975598 @default.
- W2951617797 cites W2194775991 @default.
- W2951617797 cites W2200599981 @default.
- W2951617797 cites W2209448324 @default.
- W2951617797 cites W2396622801 @default.
- W2951617797 cites W2412782625 @default.
- W2951617797 cites W2519517723 @default.
- W2951617797 cites W2557406251 @default.
- W2951617797 cites W2557728737 @default.
- W2951617797 cites W2557889580 @default.
- W2951617797 cites W2559545830 @default.
- W2951617797 cites W2560023338 @default.
- W2951617797 cites W2737258237 @default.
- W2951617797 cites W2776622059 @default.
- W2951617797 cites W2949117887 @default.
- W2951617797 cites W2949842255 @default.
- W2951617797 cites W2950898568 @default.
- W2951617797 cites W2953106684 @default.
- W2951617797 cites W2953118818 @default.
- W2951617797 cites W2953139137 @default.
- W2951617797 cites W2953203558 @default.
- W2951617797 cites W2953236957 @default.
- W2951617797 cites W2963084599 @default.
- W2951617797 cites W2963248296 @default.
- W2951617797 cites W2963840672 @default.
- W2951617797 cites W2963971305 @default.
- W2951617797 cites W2964211168 @default.
- W2951617797 cites W2964288706 @default.
- W2951617797 hasPublicationYear "2018" @default.
- W2951617797 type Work @default.
- W2951617797 sameAs 2951617797 @default.
- W2951617797 citedByCount "5" @default.
- W2951617797 countsByYear W29516177972018 @default.
- W2951617797 countsByYear W29516177972019 @default.
- W2951617797 countsByYear W29516177972020 @default.
- W2951617797 crossrefType "posted-content" @default.
- W2951617797 hasAuthorship W2951617797A5030331254 @default.
- W2951617797 hasAuthorship W2951617797A5043244870 @default.
- W2951617797 hasAuthorship W2951617797A5070642269 @default.
- W2951617797 hasAuthorship W2951617797A5078831654 @default.
- W2951617797 hasConcept C111919701 @default.
- W2951617797 hasConcept C118505674 @default.
- W2951617797 hasConcept C119857082 @default.
- W2951617797 hasConcept C126042441 @default.
- W2951617797 hasConcept C132525143 @default.
- W2951617797 hasConcept C134306372 @default.
- W2951617797 hasConcept C153180895 @default.
- W2951617797 hasConcept C154945302 @default.
- W2951617797 hasConcept C177148314 @default.
- W2951617797 hasConcept C190694206 @default.
- W2951617797 hasConcept C2776321320 @default.
- W2951617797 hasConcept C33923547 @default.
- W2951617797 hasConcept C41008148 @default.
- W2951617797 hasConcept C76155785 @default.
- W2951617797 hasConcept C80444323 @default.
- W2951617797 hasConcept C89600930 @default.
- W2951617797 hasConceptScore W2951617797C111919701 @default.
- W2951617797 hasConceptScore W2951617797C118505674 @default.
- W2951617797 hasConceptScore W2951617797C119857082 @default.
- W2951617797 hasConceptScore W2951617797C126042441 @default.
- W2951617797 hasConceptScore W2951617797C132525143 @default.
- W2951617797 hasConceptScore W2951617797C134306372 @default.
- W2951617797 hasConceptScore W2951617797C153180895 @default.
- W2951617797 hasConceptScore W2951617797C154945302 @default.
- W2951617797 hasConceptScore W2951617797C177148314 @default.
- W2951617797 hasConceptScore W2951617797C190694206 @default.
- W2951617797 hasConceptScore W2951617797C2776321320 @default.
- W2951617797 hasConceptScore W2951617797C33923547 @default.
- W2951617797 hasConceptScore W2951617797C41008148 @default.
- W2951617797 hasConceptScore W2951617797C76155785 @default.
- W2951617797 hasConceptScore W2951617797C80444323 @default.
- W2951617797 hasConceptScore W2951617797C89600930 @default.
- W2951617797 hasLocation W29516177971 @default.
- W2951617797 hasOpenAccess W2951617797 @default.
- W2951617797 hasPrimaryLocation W29516177971 @default.
- W2951617797 hasRelatedWork W1861492603 @default.
- W2951617797 hasRelatedWork W1901129140 @default.
- W2951617797 hasRelatedWork W1903029394 @default.
- W2951617797 hasRelatedWork W2031489346 @default.