arxiv.org/abs/2001.04388
Preview meta tags from the arxiv.org website.
Linked Hostnames
25- 24 links toarxiv.org
- 12 links toinfo.arxiv.org
- 6 links todblp.uni-trier.de
- 2 links tosubscribe.sorryapp.com
- 1 link toapi.semanticscholar.org
- 1 link tocore.ac.uk
- 1 link todagshub.com
- 1 link todoi.org
Thumbnail
Search Engine Appearance
RoutedFusion: Learning Real-time Depth Map Fusion
The efficient fusion of depth maps is a key part of most state-of-the-art 3D reconstruction methods. Besides requiring high accuracy, these depth fusion methods need to be scalable and real-time capable. To this end, we present a novel real-time capable machine learning-based method for depth map fusion. Similar to the seminal depth map fusion approach by Curless and Levoy, we only update a local group of voxels to ensure real-time capability. Instead of a simple linear fusion of depth information, we propose a neural network that predicts non-linear updates to better account for typical fusion errors. Our network is composed of a 2D depth routing network and a 3D depth fusion network which efficiently handle sensor-specific noise and outliers. This is especially useful for surface edges and thin objects for which the original approach suffers from thickening artifacts. Our method outperforms the traditional fusion approach and related learned approaches on both synthetic and real data. We demonstrate the performance of our method in reconstructing fine geometric details from noise and outlier contaminated data on various scenes.
Bing
RoutedFusion: Learning Real-time Depth Map Fusion
The efficient fusion of depth maps is a key part of most state-of-the-art 3D reconstruction methods. Besides requiring high accuracy, these depth fusion methods need to be scalable and real-time capable. To this end, we present a novel real-time capable machine learning-based method for depth map fusion. Similar to the seminal depth map fusion approach by Curless and Levoy, we only update a local group of voxels to ensure real-time capability. Instead of a simple linear fusion of depth information, we propose a neural network that predicts non-linear updates to better account for typical fusion errors. Our network is composed of a 2D depth routing network and a 3D depth fusion network which efficiently handle sensor-specific noise and outliers. This is especially useful for surface edges and thin objects for which the original approach suffers from thickening artifacts. Our method outperforms the traditional fusion approach and related learned approaches on both synthetic and real data. We demonstrate the performance of our method in reconstructing fine geometric details from noise and outlier contaminated data on various scenes.
DuckDuckGo
RoutedFusion: Learning Real-time Depth Map Fusion
The efficient fusion of depth maps is a key part of most state-of-the-art 3D reconstruction methods. Besides requiring high accuracy, these depth fusion methods need to be scalable and real-time capable. To this end, we present a novel real-time capable machine learning-based method for depth map fusion. Similar to the seminal depth map fusion approach by Curless and Levoy, we only update a local group of voxels to ensure real-time capability. Instead of a simple linear fusion of depth information, we propose a neural network that predicts non-linear updates to better account for typical fusion errors. Our network is composed of a 2D depth routing network and a 3D depth fusion network which efficiently handle sensor-specific noise and outliers. This is especially useful for surface edges and thin objects for which the original approach suffers from thickening artifacts. Our method outperforms the traditional fusion approach and related learned approaches on both synthetic and real data. We demonstrate the performance of our method in reconstructing fine geometric details from noise and outlier contaminated data on various scenes.
General Meta Tags
18- title[2001.04388] RoutedFusion: Learning Real-time Depth Map Fusion
- titleopen search
- titleopen navigation menu
- titlecontact arXiv
- titlesubscribe to arXiv mailings
Open Graph Meta Tags
10- og:typewebsite
- og:site_namearXiv.org
- og:titleRoutedFusion: Learning Real-time Depth Map Fusion
- og:urlhttps://arxiv.org/abs/2001.04388v2
- og:image/static/browse/0.3.4/images/arxiv-logo-fb.png
Twitter Meta Tags
6- twitter:site@arxiv
- twitter:cardsummary
- twitter:titleRoutedFusion: Learning Real-time Depth Map Fusion
- twitter:descriptionThe efficient fusion of depth maps is a key part of most state-of-the-art 3D reconstruction methods. Besides requiring high accuracy, these depth fusion methods need to be scalable and real-time...
- twitter:imagehttps://static.arxiv.org/icons/twitter/arxiv-logo-twitter-square.png
Link Tags
12- apple-touch-icon/static/browse/0.3.4/images/icons/apple-touch-icon.png
- canonical/abs/2001.04388
- icon/static/browse/0.3.4/images/icons/favicon-32x32.png
- icon/static/browse/0.3.4/images/icons/favicon-16x16.png
- manifest/static/browse/0.3.4/images/icons/site.webmanifest
Links
65- http://arxiv.org/licenses/nonexclusive-distrib/1.0
- http://gotit.pub/faq
- http://www.bibsonomy.org/BibtexHandler?requTask=upload&url=https://arxiv.org/abs/2001.04388&description=RoutedFusion: Learning Real-time Depth Map Fusion
- https://api.semanticscholar.org/arXiv:2001.04388
- https://arxiv.org