Matches in SemOpenAlex for { <https://semopenalex.org/work/W4210528082> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W4210528082 endingPage "3865" @default.
- W4210528082 startingPage "3858" @default.
- W4210528082 abstract "Global and local relational reasoning enable scene understanding models to perform human-like scene analysis and understanding. Scene understanding enables better semantic segmentation and object-to-object interaction detection. In the medical domain, a robust surgical scene understanding model allows the automation of surgical skill evaluation, real-time monitoring of surgeon’s performance and post-surgical analysis. This letter introduces a globally-reasoned multi-task surgical scene understanding model capable of performing instrument segmentation and tool-tissue interaction detection. Here, we incorporate global relational reasoning in the latent interaction space and introduce multi-scale local (neighborhood) reasoning in the coordinate space to improve segmentation. Utilizing the multi-task model setup, the performance of the visual-semantic graph attention network in interaction detection is further enhanced through global reasoning. The global interaction space features from the segmentation module are introduced into the graph network, allowing it to detect interactions based on both node-to-node and global interaction reasoning. Our model reduces the computation cost compared to running two independent single-task models by sharing common modules, which is indispensable for practical applications. Using a sequential optimization technique, the proposed multi-task model outperforms other state-of-the-art single-task models on the MICCAI endoscopic vision challenge 2018 dataset. Additionally, we also observe the performance of the multi-task model when trained using the knowledge distillation technique. The official code implementation is made available in GitHub." @default.
- W4210528082 created "2022-02-08" @default.
- W4210528082 creator A5032340829 @default.
- W4210528082 creator A5040765329 @default.
- W4210528082 creator A5065115323 @default.
- W4210528082 creator A5066653567 @default.
- W4210528082 date "2022-04-01" @default.
- W4210528082 modified "2023-10-04" @default.
- W4210528082 title "Global-Reasoned Multi-Task Learning Model for Surgical Scene Understanding" @default.
- W4210528082 cites W1965555277 @default.
- W4210528082 cites W2194775991 @default.
- W4210528082 cites W2339712187 @default.
- W4210528082 cites W2560023338 @default.
- W4210528082 cites W2792767783 @default.
- W4210528082 cites W2963091558 @default.
- W4210528082 cites W2963097937 @default.
- W4210528082 cites W2963150697 @default.
- W4210528082 cites W2963319519 @default.
- W4210528082 cites W2979328438 @default.
- W4210528082 cites W2980225217 @default.
- W4210528082 cites W2981462961 @default.
- W4210528082 cites W2981919877 @default.
- W4210528082 cites W3011409570 @default.
- W4210528082 cites W3094926883 @default.
- W4210528082 cites W3096104140 @default.
- W4210528082 cites W3098152091 @default.
- W4210528082 cites W3105636206 @default.
- W4210528082 cites W3128096387 @default.
- W4210528082 doi "https://doi.org/10.1109/lra.2022.3146544" @default.
- W4210528082 hasPublicationYear "2022" @default.
- W4210528082 type Work @default.
- W4210528082 citedByCount "10" @default.
- W4210528082 countsByYear W42105280822022 @default.
- W4210528082 countsByYear W42105280822023 @default.
- W4210528082 crossrefType "journal-article" @default.
- W4210528082 hasAuthorship W4210528082A5032340829 @default.
- W4210528082 hasAuthorship W4210528082A5040765329 @default.
- W4210528082 hasAuthorship W4210528082A5065115323 @default.
- W4210528082 hasAuthorship W4210528082A5066653567 @default.
- W4210528082 hasBestOaLocation W42105280822 @default.
- W4210528082 hasConcept C107457646 @default.
- W4210528082 hasConcept C119857082 @default.
- W4210528082 hasConcept C127413603 @default.
- W4210528082 hasConcept C132525143 @default.
- W4210528082 hasConcept C154945302 @default.
- W4210528082 hasConcept C162324750 @default.
- W4210528082 hasConcept C187736073 @default.
- W4210528082 hasConcept C2780451532 @default.
- W4210528082 hasConcept C31972630 @default.
- W4210528082 hasConcept C41008148 @default.
- W4210528082 hasConcept C62611344 @default.
- W4210528082 hasConcept C66938386 @default.
- W4210528082 hasConcept C80444323 @default.
- W4210528082 hasConcept C89600930 @default.
- W4210528082 hasConceptScore W4210528082C107457646 @default.
- W4210528082 hasConceptScore W4210528082C119857082 @default.
- W4210528082 hasConceptScore W4210528082C127413603 @default.
- W4210528082 hasConceptScore W4210528082C132525143 @default.
- W4210528082 hasConceptScore W4210528082C154945302 @default.
- W4210528082 hasConceptScore W4210528082C162324750 @default.
- W4210528082 hasConceptScore W4210528082C187736073 @default.
- W4210528082 hasConceptScore W4210528082C2780451532 @default.
- W4210528082 hasConceptScore W4210528082C31972630 @default.
- W4210528082 hasConceptScore W4210528082C41008148 @default.
- W4210528082 hasConceptScore W4210528082C62611344 @default.
- W4210528082 hasConceptScore W4210528082C66938386 @default.
- W4210528082 hasConceptScore W4210528082C80444323 @default.
- W4210528082 hasConceptScore W4210528082C89600930 @default.
- W4210528082 hasFunder F4320322942 @default.
- W4210528082 hasFunder F4320335138 @default.
- W4210528082 hasIssue "2" @default.
- W4210528082 hasLocation W42105280821 @default.
- W4210528082 hasLocation W42105280822 @default.
- W4210528082 hasLocation W42105280823 @default.
- W4210528082 hasOpenAccess W4210528082 @default.
- W4210528082 hasPrimaryLocation W42105280821 @default.
- W4210528082 hasRelatedWork W1669643531 @default.
- W4210528082 hasRelatedWork W1982826852 @default.
- W4210528082 hasRelatedWork W2005437358 @default.
- W4210528082 hasRelatedWork W2008656436 @default.
- W4210528082 hasRelatedWork W2023558673 @default.
- W4210528082 hasRelatedWork W2110230079 @default.
- W4210528082 hasRelatedWork W2134924024 @default.
- W4210528082 hasRelatedWork W2517104666 @default.
- W4210528082 hasRelatedWork W2613186388 @default.
- W4210528082 hasRelatedWork W1967061043 @default.
- W4210528082 hasVolume "7" @default.
- W4210528082 isParatext "false" @default.
- W4210528082 isRetracted "false" @default.
- W4210528082 workType "article" @default.