Matches in SemOpenAlex for { <https://semopenalex.org/work/W2900510798> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W2900510798 endingPage "68" @default.
- W2900510798 startingPage "53" @default.
- W2900510798 abstract "As 360(^{circ }) cameras become prevalent in many autonomous systems (e.g., self-driving cars and drones), efficient 360(^{circ }) perception becomes more and more important. We propose a novel self-supervised learning approach for predicting the omnidirectional depth and camera motion from a 360(^{circ }) video. In particular, starting from the SfMLearner, which is designed for cameras with normal field-of-view, we introduce three key features to process 360(^{circ }) images efficiently. Firstly, we convert each image from equirectangular projection to cubic projection in order to avoid image distortion. In each network layer, we use Cube Padding (CP), which pads intermediate features from adjacent faces, to avoid image boundaries. Secondly, we propose a novel “spherical” photometric consistency constraint on the whole viewing sphere. In this way, no pixel will be projected outside the image boundary which typically happens in images with normal field-of-view. Finally, rather than naively estimating six independent camera motions (i.e., naively applying SfM-Learner to each face on a cube), we propose a novel camera pose consistency loss to ensure the estimated camera motions reaching consensus. To train and evaluate our approach, we collect a new PanoSUNCG dataset containing a large amount of 360(^{circ }) videos with groundtruth depth and camera motion. Our approach achieves state-of-the-art depth prediction and camera motion estimation on PanoSUNCG with faster inference speed comparing to equirectangular. In real-world indoor videos, our approach can also achieve qualitatively reasonable depth prediction by acquiring model pre-trained on PanoSUNCG." @default.
- W2900510798 created "2018-11-29" @default.
- W2900510798 creator A5004081793 @default.
- W2900510798 creator A5008331570 @default.
- W2900510798 creator A5014665083 @default.
- W2900510798 creator A5015668000 @default.
- W2900510798 creator A5055004883 @default.
- W2900510798 creator A5056045120 @default.
- W2900510798 creator A5089070900 @default.
- W2900510798 creator A5091891363 @default.
- W2900510798 date "2019-01-01" @default.
- W2900510798 modified "2023-10-18" @default.
- W2900510798 title "Self-supervised Learning of Depth and Camera Motion from 360 $$^circ $$ Videos" @default.
- W2900510798 cites W1776042733 @default.
- W2900510798 cites W1989310712 @default.
- W2900510798 cites W2044542697 @default.
- W2900510798 cites W2146782207 @default.
- W2900510798 cites W2202251471 @default.
- W2900510798 cites W2300779272 @default.
- W2900510798 cites W2313644479 @default.
- W2900510798 cites W2520525508 @default.
- W2900510798 cites W2520707372 @default.
- W2900510798 cites W2539774955 @default.
- W2900510798 cites W2557465155 @default.
- W2900510798 cites W2580748890 @default.
- W2900510798 cites W2593745480 @default.
- W2900510798 cites W2606794968 @default.
- W2900510798 cites W2609883120 @default.
- W2900510798 cites W2611836059 @default.
- W2900510798 cites W2744749505 @default.
- W2900510798 cites W2798665861 @default.
- W2900510798 cites W2963012943 @default.
- W2900510798 cites W2963149945 @default.
- W2900510798 cites W2963339238 @default.
- W2900510798 cites W2963591054 @default.
- W2900510798 cites W2963654727 @default.
- W2900510798 cites W2963906250 @default.
- W2900510798 cites W3100388886 @default.
- W2900510798 cites W612478963 @default.
- W2900510798 doi "https://doi.org/10.1007/978-3-030-20873-8_4" @default.
- W2900510798 hasPublicationYear "2019" @default.
- W2900510798 type Work @default.
- W2900510798 sameAs 2900510798 @default.
- W2900510798 citedByCount "24" @default.
- W2900510798 countsByYear W29005107982019 @default.
- W2900510798 countsByYear W29005107982020 @default.
- W2900510798 countsByYear W29005107982021 @default.
- W2900510798 countsByYear W29005107982022 @default.
- W2900510798 countsByYear W29005107982023 @default.
- W2900510798 crossrefType "book-chapter" @default.
- W2900510798 hasAuthorship W2900510798A5004081793 @default.
- W2900510798 hasAuthorship W2900510798A5008331570 @default.
- W2900510798 hasAuthorship W2900510798A5014665083 @default.
- W2900510798 hasAuthorship W2900510798A5015668000 @default.
- W2900510798 hasAuthorship W2900510798A5055004883 @default.
- W2900510798 hasAuthorship W2900510798A5056045120 @default.
- W2900510798 hasAuthorship W2900510798A5089070900 @default.
- W2900510798 hasAuthorship W2900510798A5091891363 @default.
- W2900510798 hasConcept C104114177 @default.
- W2900510798 hasConcept C121684516 @default.
- W2900510798 hasConcept C154945302 @default.
- W2900510798 hasConcept C31972630 @default.
- W2900510798 hasConcept C41008148 @default.
- W2900510798 hasConceptScore W2900510798C104114177 @default.
- W2900510798 hasConceptScore W2900510798C121684516 @default.
- W2900510798 hasConceptScore W2900510798C154945302 @default.
- W2900510798 hasConceptScore W2900510798C31972630 @default.
- W2900510798 hasConceptScore W2900510798C41008148 @default.
- W2900510798 hasLocation W29005107981 @default.
- W2900510798 hasOpenAccess W2900510798 @default.
- W2900510798 hasPrimaryLocation W29005107981 @default.
- W2900510798 hasRelatedWork W2029249305 @default.
- W2900510798 hasRelatedWork W2105769806 @default.
- W2900510798 hasRelatedWork W2115571026 @default.
- W2900510798 hasRelatedWork W2144043954 @default.
- W2900510798 hasRelatedWork W2511137960 @default.
- W2900510798 hasRelatedWork W2604231787 @default.
- W2900510798 hasRelatedWork W2610014769 @default.
- W2900510798 hasRelatedWork W2687972263 @default.
- W2900510798 hasRelatedWork W3214088465 @default.
- W2900510798 hasRelatedWork W4321142835 @default.
- W2900510798 isParatext "false" @default.
- W2900510798 isRetracted "false" @default.
- W2900510798 magId "2900510798" @default.
- W2900510798 workType "book-chapter" @default.