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

Xiaoge Song, Danhuai Zhao, Wei Zhu, Kang Zheng, Tong Lu

Browse the full ICASSP paper archive.