Kim, H., Kang, J., Heo, M., Hwang, S., Oh, S. W., & Kim, S. J. (2023). VISAGE: Video Instance Segmentation with Appearance-Guided Enhancement.
Citazione stile Chigago Style (17a edizione)Kim, Hanjung, Jaehyun Kang, Miran Heo, Sukjun Hwang, Seoung Wug Oh, e Seon Joo Kim. VISAGE: Video Instance Segmentation with Appearance-Guided Enhancement. 2023.
Citatione MLA (9a ed.)Kim, Hanjung, et al. VISAGE: Video Instance Segmentation with Appearance-Guided Enhancement. 2023.
Attenzione: Queste citazioni potrebbero non essere precise al 100%.