ELECT keeps story characters consistent without slowing down rendering
ELECT: Settling Identity Before Rendering for Scalable Story Visualization
Computer Vision and Pattern RecognitionArtificial IntelligenceMachine Learning
Summary
When creating pictures for long stories without training on example images, it’s important to keep characters looking the same across all scenes. Existing methods slow down or use a lot of memory as the story gets longer because they link each image to others. The authors propose ELECT, which decides which characters go where before making the final image, so the time and memory needed stay almost the same no matter how many scenes there are. They tested ELECT on a benchmark and found it keeps characters consistent and shows text better while being much faster and using less memory than previous methods.
What this means in practice
- •For animation studios: Generate consistent character images across many story scenes efficiently without training on reference images.
- •For game developers: Create long narrative visuals that maintain character identity while keeping rendering time and memory use low.
Authors
Mehrajul Abadin Miraj, Md. Mahmudul Hasan, Meherun Farzana, Aniket Joarder, Mahmudul Hasan, Abir Chakraborty Partha, Md. Ahasanul Alam, Redwan Ahmed Rizvee, Md. Fahim Arefin, Md Mahmudur Rahman, Md. Tanvir Alam, Md Mosaddek Khan
Abstract
Training-free story visualization without reference images must preserve recurring entities across frames while scaling to long narratives. Existing methods typically maintain consistency by conditioning each frame on other frames of the same story, coupling generation time, VRAM, or both to story length. We introduce ELECT, which instead settles cross-frame identity before final rendering through a language-guided scene-entity mapping and stationary anchors elected from the model's own drafts. This makes the per-frame cost of the final rendering pass independent of story length. Joint attention exceeds 96 GB of VRAM at 32 scenes, whereas ELECT remains at 8.3 s per frame and under 20 GB VRAM from 10 to 1000 scenes. On the Story50 long-story benchmark, ELECT reaches 0.833 average Character Consistency Similarity (aCCS), compared with 0.826 for ten-round refinement, at one seventh of its generation cost, with lower text alignment (0.327 vs. 0.340). At 1000 scenes, to our knowledge, the first evaluation of training-free story visualization at this length, ELECT exceeds ten-round refinement by 0.017 aCCS and the nearest flat-cost baseline by 0.091, while also surpassing refinement in text alignment at one twenty-fourth of its generation time. These results show that long-horizon consistency need not require increasingly expensive inter-frame conditioning as stories grow.