Matches in SemOpenAlex for { <https://semopenalex.org/work/W2950405180> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W2950405180 abstract "Indoor scene understanding is central to applications such as robot navigation and human companion assistance. Over the last years, data-driven deep neural networks have outperformed many traditional approaches thanks to their representation learning capabilities. One of the bottlenecks in training for better representations is the amount of available per-pixel ground truth data that is required for core scene understanding tasks such as semantic segmentation, normal prediction, and object edge detection. To address this problem, a number of works proposed using synthetic data. However, a systematic study of how such synthetic data is generated is missing. In this work, we introduce a large-scale synthetic dataset with 400K physically-based rendered images from 45K realistic 3D indoor scenes. We study the effects of rendering methods and scene lighting on training for three computer vision tasks: surface normal prediction, semantic segmentation, and object boundary detection. This study provides insights into the best practices for training with synthetic data (more realistic rendering is worth it) and shows that pretraining with our new synthetic dataset can improve results beyond the current state of the art on all three tasks." @default.
- W2950405180 created "2019-06-27" @default.
- W2950405180 creator A5004644695 @default.
- W2950405180 creator A5028817790 @default.
- W2950405180 creator A5044267273 @default.
- W2950405180 creator A5058136382 @default.
- W2950405180 creator A5063499488 @default.
- W2950405180 creator A5079619886 @default.
- W2950405180 creator A5091765070 @default.
- W2950405180 date "2016-12-21" @default.
- W2950405180 modified "2023-10-16" @default.
- W2950405180 title "Physically-Based Rendering for Indoor Scene Understanding Using Convolutional Neural Networks" @default.
- W2950405180 cites W125693051 @default.
- W2950405180 cites W1686810756 @default.
- W2950405180 cites W1949568868 @default.
- W2950405180 cites W2008693968 @default.
- W2950405180 cites W2036242214 @default.
- W2950405180 cites W2058535340 @default.
- W2950405180 cites W2067912884 @default.
- W2950405180 cites W2190691619 @default.
- W2950405180 cites W2283234189 @default.
- W2950405180 cites W2286929393 @default.
- W2950405180 cites W2297454107 @default.
- W2950405180 cites W2431874326 @default.
- W2950405180 cites W2519379752 @default.
- W2950405180 cites W2607333215 @default.
- W2950405180 cites W2949366274 @default.
- W2950405180 cites W2950025457 @default.
- W2950405180 cites W2951713345 @default.
- W2950405180 cites W2951747365 @default.
- W2950405180 cites W2952632681 @default.
- W2950405180 cites W2952787450 @default.
- W2950405180 cites W2953221084 @default.
- W2950405180 cites W337610345 @default.
- W2950405180 doi "https://doi.org/10.48550/arxiv.1612.07429" @default.
- W2950405180 hasPublicationYear "2016" @default.
- W2950405180 type Work @default.
- W2950405180 sameAs 2950405180 @default.
- W2950405180 citedByCount "3" @default.
- W2950405180 countsByYear W29504051802017 @default.
- W2950405180 countsByYear W29504051802021 @default.
- W2950405180 crossrefType "posted-content" @default.
- W2950405180 hasAuthorship W2950405180A5004644695 @default.
- W2950405180 hasAuthorship W2950405180A5028817790 @default.
- W2950405180 hasAuthorship W2950405180A5044267273 @default.
- W2950405180 hasAuthorship W2950405180A5058136382 @default.
- W2950405180 hasAuthorship W2950405180A5063499488 @default.
- W2950405180 hasAuthorship W2950405180A5079619886 @default.
- W2950405180 hasAuthorship W2950405180A5091765070 @default.
- W2950405180 hasBestOaLocation W29504051801 @default.
- W2950405180 hasConcept C108583219 @default.
- W2950405180 hasConcept C146849305 @default.
- W2950405180 hasConcept C154945302 @default.
- W2950405180 hasConcept C160633673 @default.
- W2950405180 hasConcept C160920958 @default.
- W2950405180 hasConcept C205711294 @default.
- W2950405180 hasConcept C2776151529 @default.
- W2950405180 hasConcept C2984842247 @default.
- W2950405180 hasConcept C31972630 @default.
- W2950405180 hasConcept C41008148 @default.
- W2950405180 hasConcept C50644808 @default.
- W2950405180 hasConcept C51632099 @default.
- W2950405180 hasConcept C81363708 @default.
- W2950405180 hasConcept C89600930 @default.
- W2950405180 hasConceptScore W2950405180C108583219 @default.
- W2950405180 hasConceptScore W2950405180C146849305 @default.
- W2950405180 hasConceptScore W2950405180C154945302 @default.
- W2950405180 hasConceptScore W2950405180C160633673 @default.
- W2950405180 hasConceptScore W2950405180C160920958 @default.
- W2950405180 hasConceptScore W2950405180C205711294 @default.
- W2950405180 hasConceptScore W2950405180C2776151529 @default.
- W2950405180 hasConceptScore W2950405180C2984842247 @default.
- W2950405180 hasConceptScore W2950405180C31972630 @default.
- W2950405180 hasConceptScore W2950405180C41008148 @default.
- W2950405180 hasConceptScore W2950405180C50644808 @default.
- W2950405180 hasConceptScore W2950405180C51632099 @default.
- W2950405180 hasConceptScore W2950405180C81363708 @default.
- W2950405180 hasConceptScore W2950405180C89600930 @default.
- W2950405180 hasLocation W29504051801 @default.
- W2950405180 hasOpenAccess W2950405180 @default.
- W2950405180 hasPrimaryLocation W29504051801 @default.
- W2950405180 hasRelatedWork W2048402902 @default.
- W2950405180 hasRelatedWork W2607572884 @default.
- W2950405180 hasRelatedWork W2739874619 @default.
- W2950405180 hasRelatedWork W2744103196 @default.
- W2950405180 hasRelatedWork W2774444957 @default.
- W2950405180 hasRelatedWork W2979303128 @default.
- W2950405180 hasRelatedWork W3102253946 @default.
- W2950405180 hasRelatedWork W3144574764 @default.
- W2950405180 hasRelatedWork W4293211451 @default.
- W2950405180 hasRelatedWork W3182299699 @default.
- W2950405180 isParatext "false" @default.
- W2950405180 isRetracted "false" @default.
- W2950405180 magId "2950405180" @default.
- W2950405180 workType "article" @default.