English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Optical Flow Estimation aims to find the 2D dense motion field between twoframes. Due to the limitation of model structures and training datasets,existing methods often rely too much on local clues and ignore the integrity ofobjects, resulting in fragmented motion estimation. Through theoreticalanalysis, we find the pre-trained large vision models are helpful in opticalflow estimation, and we notice that the recently famous Segment Anything Model(SAM) demonstrates a strong ability to segment complete objects, which issuitable for solving the fragmentation problem. We thus propose a solution toembed the frozen SAM image encoder into FlowFormer to enhance objectperception. To address the challenge of in-depth utilizing SAM innon-segmentation tasks like optical flow estimation, we propose an Optical FlowTask-Specific Adaption scheme, including a Context Fusion Module to fuse theSAM encoder with the optical flow context encoder, and a Context AdaptionModule to adapt the SAM features for optical flow task with LearnedTask-Specific Embedding. Our proposed SAMFlow model reaches 0.86/2.10clean/final EPE and 3.55/12.32 EPE/F1-all on Sintel and KITTI-15 training set,surpassing Flowformer by 8.5%/9.9% and 13.2%/16.3%. Furthermore, our modelachieves state-of-the-art performance on the Sintel and KITTI-15 benchmarks,ranking #1 among all two-frame methods on Sintel clean pass.

SAMFlow: Eliminating Any Fragmentation in Optical Flow with Segment Anything Model