Conditional Convolutions for End-to-End Single-Stage Video Text Detection

Song, Xingrui; Zhao, Danhuai; Zhu, W. J.; Kang, Zheng; Lu, Tong

doi:10.1109/icassp49660.2025.10889372

ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2025

DOI: 10.1109/icassp49660.2025.10889372

|Get access via publisher |Cite

Conditional Convolutions for End-to-End Single-Stage Video Text Detection

Xingrui Song

Danhuai Zhao

W. J. Zhu

et al.

Help me understand this report

This publication either has no citations yet, or we are still processing them

Set email alert for when this publication receives citations?

See others like this or search for similar articles

Scite is an AI-powered research tool that helps researchers better discover and evaluate scientific literature through Smart Citations—a revolutionary system that shows whether articles support, contrast, or simply mention a given claim. Founded in 2018, and now part of Research Solutions, Scite has indexed over 1.3 billion citations and partnered with more than 30 major publishers to provide researchers with unparalleled access to scientific literature. With its Scite Assistant, Smart Citation Index, and advanced search capabilities, the platform addresses critical challenges such as information overload and research reproducibility. Trusted by two million active users worldwide, Scite is reshaping how researchers interact with scholarly content—building ethical, transparent AI tools that support rigorous, copyright-compliant research.