Matches in SemOpenAlex for { <https://semopenalex.org/work/W4288804593> ?p ?o ?g. }
Showing items 1 to 61 of
61
with 100 items per page.
- W4288804593 abstract "While Video Instance Segmentation (VIS) has seen rapid progress, current approaches struggle to predict high-quality masks with accurate boundary details. Moreover, the predicted segmentations often fluctuate over time, suggesting that temporal consistency cues are neglected or not fully utilized. In this paper, we set out to tackle these issues, with the aim of achieving highly detailed and more temporally stable mask predictions for VIS. We first propose the Video Mask Transfiner (VMT) method, capable of leveraging fine-grained high-resolution features thanks to a highly efficient video transformer structure. Our VMT detects and groups sparse error-prone spatio-temporal regions of each tracklet in the video segment, which are then refined using both local and instance-level cues. Second, we identify that the coarse boundary annotations of the popular YouTube-VIS dataset constitute a major limiting factor. Based on our VMT architecture, we therefore design an automated annotation refinement approach by iterative training and self-correction. To benchmark high-quality mask predictions for VIS, we introduce the HQ-YTVIS dataset, consisting of a manually re-annotated test set and our automatically refined training data. We compare VMT with the most recent state-of-the-art methods on the HQ-YTVIS, as well as the Youtube-VIS, OVIS and BDD100K MOTS benchmarks. Experimental results clearly demonstrate the efficacy and effectiveness of our method on segmenting complex and dynamic objects, by capturing precise details." @default.
- W4288804593 created "2022-07-30" @default.
- W4288804593 creator A5001987788 @default.
- W4288804593 creator A5036631624 @default.
- W4288804593 creator A5062566088 @default.
- W4288804593 creator A5067487326 @default.
- W4288804593 creator A5068158266 @default.
- W4288804593 creator A5075499930 @default.
- W4288804593 date "2022-07-28" @default.
- W4288804593 modified "2023-09-29" @default.
- W4288804593 title "Video Mask Transfiner for High-Quality Video Instance Segmentation" @default.
- W4288804593 doi "https://doi.org/10.48550/arxiv.2207.14012" @default.
- W4288804593 hasPublicationYear "2022" @default.
- W4288804593 type Work @default.
- W4288804593 citedByCount "0" @default.
- W4288804593 crossrefType "posted-content" @default.
- W4288804593 hasAuthorship W4288804593A5001987788 @default.
- W4288804593 hasAuthorship W4288804593A5036631624 @default.
- W4288804593 hasAuthorship W4288804593A5062566088 @default.
- W4288804593 hasAuthorship W4288804593A5067487326 @default.
- W4288804593 hasAuthorship W4288804593A5068158266 @default.
- W4288804593 hasAuthorship W4288804593A5075499930 @default.
- W4288804593 hasBestOaLocation W42888045931 @default.
- W4288804593 hasConcept C13280743 @default.
- W4288804593 hasConcept C153180895 @default.
- W4288804593 hasConcept C154945302 @default.
- W4288804593 hasConcept C177264268 @default.
- W4288804593 hasConcept C185798385 @default.
- W4288804593 hasConcept C199360897 @default.
- W4288804593 hasConcept C205649164 @default.
- W4288804593 hasConcept C2776321320 @default.
- W4288804593 hasConcept C31972630 @default.
- W4288804593 hasConcept C41008148 @default.
- W4288804593 hasConcept C89600930 @default.
- W4288804593 hasConceptScore W4288804593C13280743 @default.
- W4288804593 hasConceptScore W4288804593C153180895 @default.
- W4288804593 hasConceptScore W4288804593C154945302 @default.
- W4288804593 hasConceptScore W4288804593C177264268 @default.
- W4288804593 hasConceptScore W4288804593C185798385 @default.
- W4288804593 hasConceptScore W4288804593C199360897 @default.
- W4288804593 hasConceptScore W4288804593C205649164 @default.
- W4288804593 hasConceptScore W4288804593C2776321320 @default.
- W4288804593 hasConceptScore W4288804593C31972630 @default.
- W4288804593 hasConceptScore W4288804593C41008148 @default.
- W4288804593 hasConceptScore W4288804593C89600930 @default.
- W4288804593 hasLocation W42888045931 @default.
- W4288804593 hasOpenAccess W4288804593 @default.
- W4288804593 hasPrimaryLocation W42888045931 @default.
- W4288804593 hasRelatedWork W10362453 @default.
- W4288804593 hasRelatedWork W10684755 @default.
- W4288804593 hasRelatedWork W13986278 @default.
- W4288804593 hasRelatedWork W14128562 @default.
- W4288804593 hasRelatedWork W3542719 @default.
- W4288804593 hasRelatedWork W4554164 @default.
- W4288804593 hasRelatedWork W6426610 @default.
- W4288804593 hasRelatedWork W7811848 @default.
- W4288804593 hasRelatedWork W9584637 @default.
- W4288804593 hasRelatedWork W9877987 @default.
- W4288804593 isParatext "false" @default.
- W4288804593 isRetracted "false" @default.
- W4288804593 workType "article" @default.