Matches in SemOpenAlex for { <https://semopenalex.org/work/W4206239741> ?p ?o ?g. }
- W4206239741 endingPage "4652" @default.
- W4206239741 startingPage "4639" @default.
- W4206239741 abstract "Optical flow estimation is a fundamental task in computer vision and image processing. Due to the difficulty in obtaining the ground truth of flow field, unsupervised learning approaches attract more and more research interests in recent years. However, despite of their good generalization capability, unsupervised optical flow methods suffer in the scenarios with large displacement, small objects, and occlusions. In this work, we propose a novel optical flow network based on decoder with multi-scale kernels. Different from previous U-Net like or pyramidal methods, we design our network based on RAFT architecture that with a 4D correlation layer and recurrent decoder. More importantly, we incorporate three novel ideas with regard to the input, information processing and output of the update units improve the performance. Firstly, we utilize various motion-related information as input to the update units. Secondly, we propose a module of multi-scale update unit. Thirdly, for the final flow up-sampling procedure, we propose an image-guided up-sampling loss to guide the learning of up-sampling masks. Our model is trained by the occlusion-aware photometric loss, edge-aware smoothness loss, self-supervised loss, and image-guided up-sampling loss. Experimental results demonstrate that our model achieves the state-of-the-art performance on both Sintel and KITTI and outperforms other unsupervised optical flow methods remarkably." @default.
- W4206239741 created "2022-01-25" @default.
- W4206239741 creator A5009765429 @default.
- W4206239741 creator A5013727612 @default.
- W4206239741 creator A5013937448 @default.
- W4206239741 creator A5058066577 @default.
- W4206239741 creator A5079412089 @default.
- W4206239741 creator A5091007948 @default.
- W4206239741 date "2022-07-01" @default.
- W4206239741 modified "2023-10-12" @default.
- W4206239741 title "MRDFlow: Unsupervised Optical Flow Estimation Network With Multi-Scale Recurrent Decoder" @default.
- W4206239741 cites W1513100184 @default.
- W4206239741 cites W1578285471 @default.
- W4206239741 cites W1674866864 @default.
- W4206239741 cites W1867429401 @default.
- W4206239741 cites W1921093919 @default.
- W4206239741 cites W1951289974 @default.
- W4206239741 cites W1991171799 @default.
- W4206239741 cites W2033959528 @default.
- W4206239741 cites W2086516725 @default.
- W4206239741 cites W2100315781 @default.
- W4206239741 cites W2115579991 @default.
- W4206239741 cites W2130657708 @default.
- W4206239741 cites W2131747574 @default.
- W4206239741 cites W2147253850 @default.
- W4206239741 cites W2150066425 @default.
- W4206239741 cites W2158919891 @default.
- W4206239741 cites W2213164203 @default.
- W4206239741 cites W2326762225 @default.
- W4206239741 cites W2506123059 @default.
- W4206239741 cites W2510924253 @default.
- W4206239741 cites W2527824482 @default.
- W4206239741 cites W2548527721 @default.
- W4206239741 cites W2560335610 @default.
- W4206239741 cites W2560474170 @default.
- W4206239741 cites W2601564443 @default.
- W4206239741 cites W2604909019 @default.
- W4206239741 cites W2623012233 @default.
- W4206239741 cites W2779003141 @default.
- W4206239741 cites W2779333428 @default.
- W4206239741 cites W2790999834 @default.
- W4206239741 cites W2894983388 @default.
- W4206239741 cites W2901940963 @default.
- W4206239741 cites W2904071629 @default.
- W4206239741 cites W2904099109 @default.
- W4206239741 cites W2904340070 @default.
- W4206239741 cites W2907214745 @default.
- W4206239741 cites W2908106554 @default.
- W4206239741 cites W2946209203 @default.
- W4206239741 cites W2949023359 @default.
- W4206239741 cites W2962771259 @default.
- W4206239741 cites W2962864875 @default.
- W4206239741 cites W2962989121 @default.
- W4206239741 cites W2963210183 @default.
- W4206239741 cites W2963483509 @default.
- W4206239741 cites W2963782415 @default.
- W4206239741 cites W2963891416 @default.
- W4206239741 cites W2964140947 @default.
- W4206239741 cites W2964156315 @default.
- W4206239741 cites W2964199361 @default.
- W4206239741 cites W2964748237 @default.
- W4206239741 cites W2966926453 @default.
- W4206239741 cites W2967043539 @default.
- W4206239741 cites W2971686478 @default.
- W4206239741 cites W2977036535 @default.
- W4206239741 cites W2979442169 @default.
- W4206239741 cites W2980116241 @default.
- W4206239741 cites W2982361806 @default.
- W4206239741 cites W2997216914 @default.
- W4206239741 cites W3001201412 @default.
- W4206239741 cites W3007097210 @default.
- W4206239741 cites W3034896357 @default.
- W4206239741 cites W3048275157 @default.
- W4206239741 cites W3048979322 @default.
- W4206239741 cites W3087584369 @default.
- W4206239741 cites W3096376821 @default.
- W4206239741 cites W3097139398 @default.
- W4206239741 cites W3100280854 @default.
- W4206239741 cites W3108086282 @default.
- W4206239741 cites W3109908659 @default.
- W4206239741 cites W3109923092 @default.
- W4206239741 cites W3127528059 @default.
- W4206239741 cites W764651262 @default.
- W4206239741 doi "https://doi.org/10.1109/tcsvt.2021.3135440" @default.
- W4206239741 hasPublicationYear "2022" @default.
- W4206239741 type Work @default.
- W4206239741 citedByCount "3" @default.
- W4206239741 countsByYear W42062397412023 @default.
- W4206239741 crossrefType "journal-article" @default.
- W4206239741 hasAuthorship W4206239741A5009765429 @default.
- W4206239741 hasAuthorship W4206239741A5013727612 @default.
- W4206239741 hasAuthorship W4206239741A5013937448 @default.
- W4206239741 hasAuthorship W4206239741A5058066577 @default.
- W4206239741 hasAuthorship W4206239741A5079412089 @default.
- W4206239741 hasAuthorship W4206239741A5091007948 @default.
- W4206239741 hasConcept C106131492 @default.
- W4206239741 hasConcept C115961682 @default.
- W4206239741 hasConcept C134306372 @default.