Table Transformer (DETR) - Table Structure Recognition is a Transformer-based object detection model trained on the PubTables1M dataset. It is designed to detect and recognize table structures, such as rows and columns, in unstructured documents. Based on DETR (DEtection TRansformer), this model employs a "normalize before" approach, where layer normalization is applied before self- and cross-attention. It enables precise table structure recognition, making it useful for document processing, table parsing, and data extraction tasks.
MIT
Sector Agnostic
Microsoft
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
Open
12/03/25 06:34:56
N.A.
0
© 2026 - Copyright AIKosh. All rights reserved.