Matches in SemOpenAlex for { <https://semopenalex.org/work/W2914872565> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W2914872565 abstract "Motion is a dominant cue in automated driving systems. Optical flow is typically computed to detect moving objects and to estimate depth using triangulation. In this paper, our motivation is to leverage the existing dense optical flow to improve the performance of semantic segmentation. To provide a systematic study, we construct four different architectures which use RGB only, flow only, RGBF concatenated and two-stream RGB + flow. We evaluate these networks on two automotive datasets namely Virtual KITTI and Cityscapes using the state-of-the-art flow estimator FlowNet v2. We also make use of the ground truth optical flow in Virtual KITTI to serve as an ideal estimator and a standard Farneback optical flow algorithm to study the effect of noise. Using the flow ground truth in Virtual KITTI, two-stream architecture achieves the best results with an improvement of 4% IoU. As expected, there is a large improvement for moving objects like trucks, vans and cars with 38%, 28% and 6% increase in IoU. FlowNet produces an improvement of 2.4% in average IoU with larger improvement in the moving objects corresponding to 26%, 11% and 5% in trucks, vans and cars. In Cityscapes, flow augmentation provided an improvement for moving objects like motorcycle and train with an increase of 17% and 7% in IoU." @default.
- W2914872565 created "2019-02-21" @default.
- W2914872565 creator A5001769065 @default.
- W2914872565 creator A5002670983 @default.
- W2914872565 creator A5012401830 @default.
- W2914872565 creator A5014764449 @default.
- W2914872565 creator A5020536138 @default.
- W2914872565 date "2019-01-01" @default.
- W2914872565 modified "2023-09-28" @default.
- W2914872565 title "Optical Flow Augmented Semantic Segmentation Networks for Automated Driving" @default.
- W2914872565 cites W1903029394 @default.
- W2914872565 cites W1904301353 @default.
- W2914872565 cites W1907877624 @default.
- W2914872565 cites W2002875082 @default.
- W2914872565 cites W2113708607 @default.
- W2914872565 cites W2517104666 @default.
- W2914872565 cites W2561585794 @default.
- W2914872565 cites W2582761847 @default.
- W2914872565 cites W2587989515 @default.
- W2914872565 cites W2608018946 @default.
- W2914872565 cites W2744014255 @default.
- W2914872565 cites W2754856369 @default.
- W2914872565 cites W2894138533 @default.
- W2914872565 cites W2949907962 @default.
- W2914872565 cites W2952186347 @default.
- W2914872565 cites W2952390197 @default.
- W2914872565 cites W2952637581 @default.
- W2914872565 cites W2953139137 @default.
- W2914872565 cites W2963881378 @default.
- W2914872565 cites W2734839544 @default.
- W2914872565 doi "https://doi.org/10.5220/0007248300002108" @default.
- W2914872565 hasPublicationYear "2019" @default.
- W2914872565 type Work @default.
- W2914872565 sameAs 2914872565 @default.
- W2914872565 citedByCount "1" @default.
- W2914872565 countsByYear W29148725652019 @default.
- W2914872565 crossrefType "proceedings-article" @default.
- W2914872565 hasAuthorship W2914872565A5001769065 @default.
- W2914872565 hasAuthorship W2914872565A5002670983 @default.
- W2914872565 hasAuthorship W2914872565A5012401830 @default.
- W2914872565 hasAuthorship W2914872565A5014764449 @default.
- W2914872565 hasAuthorship W2914872565A5020536138 @default.
- W2914872565 hasBestOaLocation W29148725652 @default.
- W2914872565 hasConcept C115961682 @default.
- W2914872565 hasConcept C124504099 @default.
- W2914872565 hasConcept C154945302 @default.
- W2914872565 hasConcept C155542232 @default.
- W2914872565 hasConcept C204321447 @default.
- W2914872565 hasConcept C31972630 @default.
- W2914872565 hasConcept C41008148 @default.
- W2914872565 hasConcept C89600930 @default.
- W2914872565 hasConceptScore W2914872565C115961682 @default.
- W2914872565 hasConceptScore W2914872565C124504099 @default.
- W2914872565 hasConceptScore W2914872565C154945302 @default.
- W2914872565 hasConceptScore W2914872565C155542232 @default.
- W2914872565 hasConceptScore W2914872565C204321447 @default.
- W2914872565 hasConceptScore W2914872565C31972630 @default.
- W2914872565 hasConceptScore W2914872565C41008148 @default.
- W2914872565 hasConceptScore W2914872565C89600930 @default.
- W2914872565 hasLocation W29148725651 @default.
- W2914872565 hasLocation W29148725652 @default.
- W2914872565 hasOpenAccess W2914872565 @default.
- W2914872565 hasPrimaryLocation W29148725651 @default.
- W2914872565 hasRelatedWork W1669643531 @default.
- W2914872565 hasRelatedWork W1721780360 @default.
- W2914872565 hasRelatedWork W1990911012 @default.
- W2914872565 hasRelatedWork W2110230079 @default.
- W2914872565 hasRelatedWork W2117664411 @default.
- W2914872565 hasRelatedWork W2117933325 @default.
- W2914872565 hasRelatedWork W2122581818 @default.
- W2914872565 hasRelatedWork W2159066190 @default.
- W2914872565 hasRelatedWork W2739874619 @default.
- W2914872565 hasRelatedWork W29292716 @default.
- W2914872565 isParatext "false" @default.
- W2914872565 isRetracted "false" @default.
- W2914872565 magId "2914872565" @default.
- W2914872565 workType "article" @default.