Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387355992> ?p ?o ?g. }
Showing items 1 to 57 of
57
with 100 items per page.
- W4387355992 abstract "Optical flow estimation is crucial for various applications in vision and robotics. As the difficulty of collecting ground truth optical flow in real-world scenarios, most of the existing methods of learning optical flow still adopt synthetic dataset for supervised training or utilize photometric consistency across temporally adjacent video frames to drive the unsupervised learning, where the former typically has issues of generalizability while the latter usually performs worse than the supervised ones. To tackle such challenges, we propose to leverage the geometric connection between optical flow estimation and stereo matching (based on the similarity upon finding pixel correspondences across images) to unify various real-world depth estimation datasets for generating supervised training data upon optical flow. Specifically, we turn the monocular depth datasets into stereo ones via synthesizing virtual disparity, thus leading to the flows along the horizontal direction; moreover, we introduce virtual camera motion into stereo data to produce additional flows along the vertical direction. Furthermore, we propose applying geometric augmentations on one image of an optical flow pair, encouraging the optical flow estimator to learn from more challenging cases. Lastly, as the optical flow maps under different geometric augmentations actually exhibit distinct characteristics, an auxiliary classifier which trains to identify the type of augmentation from the appearance of the flow map is utilized to further enhance the learning of the optical flow estimator. Our proposed method is general and is not tied to any particular flow estimator, where extensive experiments based on various datasets and optical flow estimation models verify its efficacy and superiority." @default.
- W4387355992 created "2023-10-05" @default.
- W4387355992 creator A5032257572 @default.
- W4387355992 creator A5038549076 @default.
- W4387355992 date "2023-10-03" @default.
- W4387355992 modified "2023-10-06" @default.
- W4387355992 title "Skin the sheep not only once: Reusing Various Depth Datasets to Drive the Learning of Optical Flow" @default.
- W4387355992 doi "https://doi.org/10.48550/arxiv.2310.01833" @default.
- W4387355992 hasPublicationYear "2023" @default.
- W4387355992 type Work @default.
- W4387355992 citedByCount "0" @default.
- W4387355992 crossrefType "posted-content" @default.
- W4387355992 hasAuthorship W4387355992A5032257572 @default.
- W4387355992 hasAuthorship W4387355992A5038549076 @default.
- W4387355992 hasBestOaLocation W43873559921 @default.
- W4387355992 hasConcept C105795698 @default.
- W4387355992 hasConcept C115961682 @default.
- W4387355992 hasConcept C136389625 @default.
- W4387355992 hasConcept C146849305 @default.
- W4387355992 hasConcept C153083717 @default.
- W4387355992 hasConcept C153180895 @default.
- W4387355992 hasConcept C154945302 @default.
- W4387355992 hasConcept C155542232 @default.
- W4387355992 hasConcept C185429906 @default.
- W4387355992 hasConcept C31972630 @default.
- W4387355992 hasConcept C33923547 @default.
- W4387355992 hasConcept C41008148 @default.
- W4387355992 hasConcept C50644808 @default.
- W4387355992 hasConceptScore W4387355992C105795698 @default.
- W4387355992 hasConceptScore W4387355992C115961682 @default.
- W4387355992 hasConceptScore W4387355992C136389625 @default.
- W4387355992 hasConceptScore W4387355992C146849305 @default.
- W4387355992 hasConceptScore W4387355992C153083717 @default.
- W4387355992 hasConceptScore W4387355992C153180895 @default.
- W4387355992 hasConceptScore W4387355992C154945302 @default.
- W4387355992 hasConceptScore W4387355992C155542232 @default.
- W4387355992 hasConceptScore W4387355992C185429906 @default.
- W4387355992 hasConceptScore W4387355992C31972630 @default.
- W4387355992 hasConceptScore W4387355992C33923547 @default.
- W4387355992 hasConceptScore W4387355992C41008148 @default.
- W4387355992 hasConceptScore W4387355992C50644808 @default.
- W4387355992 hasLocation W43873559921 @default.
- W4387355992 hasOpenAccess W4387355992 @default.
- W4387355992 hasPrimaryLocation W43873559921 @default.
- W4387355992 hasRelatedWork W2012410061 @default.
- W4387355992 hasRelatedWork W2063823869 @default.
- W4387355992 hasRelatedWork W2798269247 @default.
- W4387355992 hasRelatedWork W2970427506 @default.
- W4387355992 hasRelatedWork W3175363083 @default.
- W4387355992 hasRelatedWork W4287880334 @default.
- W4387355992 hasRelatedWork W4295532600 @default.
- W4387355992 hasRelatedWork W4310825149 @default.
- W4387355992 hasRelatedWork W4366700029 @default.
- W4387355992 hasRelatedWork W4386076228 @default.
- W4387355992 isParatext "false" @default.
- W4387355992 isRetracted "false" @default.
- W4387355992 workType "article" @default.