eprintid: 10199974 rev_number: 6 eprint_status: archive userid: 699 dir: disk0/10/19/99/74 datestamp: 2024-11-11 11:16:51 lastmod: 2024-11-11 11:16:51 status_changed: 2024-11-11 11:16:51 type: proceedings_section metadata_visibility: show sword_depositor: 699 creators_name: Song, Tianyi creators_name: Cao, Jiuxin creators_name: Wang, Kun creators_name: Liu, Bo creators_name: Zhang, Xiaofeng title: Causal-Story: Local Causal Attention Utilizing Parameter-Efficient Tuning for Visual Story Synthesis ispublished: pub divisions: UCL divisions: B04 divisions: F48 keywords: Training, Image quality, Visualization, Coherence, Signal processing, Acoustics, Speech processing note: This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions. abstract: The excellent text-to-image synthesis capability of diffusion models has driven progress in synthesizing coherent visual stories. The current state-of-the-art method combines the features of historical captions, historical frames, and the current captions as conditions for generating the current frame. However, this method treats each historical frame and caption as the same contribution. It connects them in order with equal weights, ignoring that not all historical conditions are associated with the generation of the current frame. To address this issue, we propose Causal-Story. This model incorporates a local causal attention mechanism that considers the causal relationship between previous captions, frames, and current captions. By assigning weights based on this relationship, Causal-Story generates the current frame, thereby improving the global consistency of story generation. We evaluated our model on the PororoSV and FlintstonesSV datasets and obtained state-of-the-art FID scores, and the generated frames also demonstrate better storytelling in visuals. date: 2024-03-18 date_type: published publisher: IEEE official_url: http://dx.doi.org/10.1109/icassp48485.2024.10446420 oa_status: green full_text_type: other language: eng primo: open primo_central: open_green verified: verified_manual elements_id: 2333361 doi: 10.1109/icassp48485.2024.10446420 lyricists_name: Song, Tianyi lyricists_id: TSONG42 actors_name: Jayawardana, Anusha actors_id: AJAYA51 actors_role: owner full_text_status: public pres_type: paper publication: ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) place_of_pub: Seoul, Korea, Republic of pagerange: 3350-3354 event_title: ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) event_dates: 14 Apr 2024 - 19 Apr 2024 book_title: ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) citation: Song, Tianyi; Cao, Jiuxin; Wang, Kun; Liu, Bo; Zhang, Xiaofeng; (2024) Causal-Story: Local Causal Attention Utilizing Parameter-Efficient Tuning for Visual Story Synthesis. In: ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). (pp. pp. 3350-3354). IEEE: Seoul, Korea, Republic of. Green open access document_url: https://discovery.ucl.ac.uk/id/eprint/10199974/1/Song_2309.09553v4.pdf