An OCR dataset for invoices, likely containing images and corresponding text annotations. The dataset includes bounding box coordinates, which suggests it is structured for training object detection models. It is hosted on Kaggle, but details on its size, origin, and creation date are not provided.
Use Cases
- Train a model to detect and extract text regions from invoice scans (inferred from domain, verify after download)
- Benchmark OCR accuracy on structured financial documents (inferred from domain, verify after download)
- Fine-tune a layout analysis model using provided bounding boxes (inferred from domain, verify after download)
Strengths
- Published on Kaggle, a platform with established data sharing infrastructure.
- The title indicates the presence of bounding box annotations, a key feature for training detection models.
Limitations
- Metadata is minimal; actual content requires verification after download.
- Row count, file formats, and data scale are unknown, which may limit suitability assessment.
- Column-level documentation is absent; field semantics must be inferred after download.