A large-scale multi-modal dataset for referring motion expression video segmentation, developed by FudanCVL. The dataset is associated with research papers published at ICCV 2023 and TPAMI 2025. The dataset page was last updated on 2025-11-27.
Use Cases
- Training models for video object segmentation based on natural language referring expressions.
- Benchmarking multi-modal AI systems that link visual motion with textual descriptions.
- Developing algorithms for segmenting objects in videos based on motion cues mentioned in the description.
Strengths
- Dataset is associated with peer-reviewed publications at major conferences (ICCV 2023) and journals (TPAMI 2025).
- Dataset page was updated on 2025 11 27, suggesting recent maintenance.
Limitations
- Description metadata is limited; actual data quality requires manual inspection after download.
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count, file formats, and license are unknown, which may limit suitability assessment.
Provenance
- Source
- FudanCVL
- Collection Method
- Likely curated for academic research on video segmentation, as indicated by the associated papers.
- Time Range
- null
- Freshness
- Last updated 2025-11-27 09:42:58; freshness should be verified.
- Geography
- null