TF-ID
Tool for recognizing tables and figures in academic literature
CommonProductProductivityAcademic literatureObject detection
TF-ID is an object detection model series created by Yifei Hu for extracting tables and figures from academic papers. These models are fine-tuned based on the microsoft/Florence-2 checkpoint, offering versions with or without title text. Their aim is to enhance the accessibility and processing efficiency of information in academic literature.
TF-ID Visit Over Time
Monthly Visits
515580771
Bounce Rate
37.20%
Page per Visit
5.8
Visit Duration
00:06:42