Character-centric Story Visualization via Visual Planning and Token Alignment

Hong Chen, Rujun Han, Te-Lin Wu, Hideki Nakayama, Nanyun Peng


Abstract
Story visualization advances the traditional text-to-image generation by enabling multiple image generation based on a complete story. This task requires machines to 1) understand long text inputs, and 2) produce a globally consistent image sequence that illustrates the contents of the story. A key challenge of consistent story visualization is to preserve characters that are essential in stories. To tackle the challenge, we propose to adapt a recent work that augments VQ-VAE with a text-to-visual-token (transformer) architecture. Specifically, we modify the text-to-visual-token module with a two-stage framework: 1) character token planning model that predicts the visual tokens for characters only; 2) visual token completion model that generates the remaining visual token sequence, which is sent to VQ-VAE for finalizing image generations. To encourage characters to appear in the images, we further train the two-stage framework with a character-token alignment objective. Extensive experiments and evaluations demonstrate that the proposed method excels at preserving characters and can produce higher quality image sequences compared with the strong baselines.
Anthology ID:
2022.emnlp-main.565
Volume:
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates
Editors:
Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
8259–8272
Language:
URL:
https://aclanthology.org/2022.emnlp-main.565
DOI:
10.18653/v1/2022.emnlp-main.565
Bibkey:
Cite (ACL):
Hong Chen, Rujun Han, Te-Lin Wu, Hideki Nakayama, and Nanyun Peng. 2022. Character-centric Story Visualization via Visual Planning and Token Alignment. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 8259–8272, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
Cite (Informal):
Character-centric Story Visualization via Visual Planning and Token Alignment (Chen et al., EMNLP 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.emnlp-main.565.pdf