Picture for Gangyan Zeng

Gangyan Zeng

Focus, Distinguish, and Prompt: Unleashing CLIP for Efficient and Flexible Scene Text Retrieval

Add code
Aug 01, 2024
Viaarxiv icon

TextBlockV2: Towards Precise-Detection-Free Scene Text Spotting with Pre-trained Language Model

Add code
Mar 15, 2024
Viaarxiv icon

Towards Escaping from Language Bias and OCR Error: Semantics-Centered Text Visual Question Answering

Add code
Mar 24, 2022
Figure 1 for Towards Escaping from Language Bias and OCR Error: Semantics-Centered Text Visual Question Answering
Figure 2 for Towards Escaping from Language Bias and OCR Error: Semantics-Centered Text Visual Question Answering
Figure 3 for Towards Escaping from Language Bias and OCR Error: Semantics-Centered Text Visual Question Answering
Figure 4 for Towards Escaping from Language Bias and OCR Error: Semantics-Centered Text Visual Question Answering
Viaarxiv icon