http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Real Scene Text Image Super-Resolution Based on Multi-Scale and Attention Fusion
Xinhua Lu,Haihai Wei,Li Ma,Qingji Xue,Yonghui Fu 한국정보처리학회 2023 Journal of information processing systems Vol.19 No.4
Plenty of works have indicated that single image super-resolution (SISR) models relying on synthetic datasetsare difficult to be applied to real scene text image super-resolution (STISR) for its more complex degradation. The up-to-date dataset for realistic STISR is called TextZoom, while the current methods trained on this datasethave not considered the effect of multi-scale features of text images. In this paper, a multi-scale and attentionfusion model for realistic STISR is proposed. The multi-scale learning mechanism is introduced to acquiresophisticated feature representations of text images; The spatial and channel attentions are introduced to capturethe local information and inter-channel interaction information of text images; At last, this paper designs amulti-scale residual attention module by skillfully fusing multi-scale learning and attention mechanisms. Theexperiments on TextZoom demonstrate that the model proposed increases scene text recognition’s (ASTER)average recognition accuracy by 1.2% compared to text super-resolution network.