YouTube-VOS
Emerging14papers using it
17HF downloads
0HF likes
2019first seen
YouTube-VOS is a benchmark dataset used to evaluate semi-supervised video object segmentation by providing a collection of video sequences with annotated object masks.
Papers using YouTube-VOS (14)
- Learning Position and Target Consistency for Memory-based Video Object
SegmentationVORNet: Spatio-temporally Consistent Video Inpainting for Object RemovalCollaborative Video Object Segmentation by Foreground-Background
IntegrationVideo Instance SegmentationTransVOS: Video Object Segmentation with TransformersCollaborative Video Object Segmentation by Multi-Scale
Foreground-Background IntegrationDiscriminative Online Learning for Fast Video Object SegmentationLearning Dynamic Network Using a Reuse Gate Function in Semi-supervised
Video Object SegmentationWeakly Supervised Few-shot Object Segmentation using Co-Attention with
Visual and Semantic EmbeddingsVideo Instance Segmentation by Instance Flow AssemblyObject Propagation via Inter-Frame Attentions for Temporally Stable
Video Instance SegmentationFlowVOS: Weakly-Supervised Visual Warping for Detail-Preserving and
Temporally Consistent Single-Shot Video Object SegmentationHierarchical Spatiotemporal Transformers for Video Object SegmentationSSTVOS: Sparse Spatiotemporal Transformers for Video Object Segmentation