A dataset likely containing pre-computed visual features for the Flickr30K image collection. The title suggests features were extracted using YOLO and Xception models, possibly for the Scene Classification 2 (SC2) task. It is hosted on Kaggle, but specific details on the number of samples, feature dimensions, and creation date are unavailable.
Use Cases
- Training or benchmarking image captioning models using pre-computed visual embeddings (inferred from domain, verify after download)
- Fine-tuning vision-language models for downstream tasks like visual question answering (inferred from domain, verify after download)
- Analyzing the effectiveness of YOLO and Xception feature extractors for scene classification (inferred from domain, verify after download)
Strengths
- Published on Kaggle, a platform with established data sharing infrastructure.
- The title indicates use of established computer vision models (YOLO, Xception) for feature extraction.
Limitations
- Metadata is minimal; actual content requires verification after download.
- Row count, file size, column definitions, and license are unknown.
- Last update date is unknown; freshness unverified.
Provenance
- Source
- null
- Collection Method
- Features likely extracted from the Flickr30K image dataset using YOLO and Xception models.
- Time Range
- null
- Freshness
- null
- Geography
- null