Kaleido: Open-Sourced Multi-Subject Reference Video Generation Model
By: Zhenxing Zhang , Jiayan Teng , Zhuoyi Yang and more
Potential Business Impact:
Makes videos of people look real from pictures.
We present Kaleido, a subject-to-video~(S2V) generation framework, which aims to synthesize subject-consistent videos conditioned on multiple reference images of target subjects. Despite recent progress in S2V generation models, existing approaches remain inadequate at maintaining multi-subject consistency and at handling background disentanglement, often resulting in lower reference fidelity and semantic drift under multi-image conditioning. These shortcomings can be attributed to several factors. Primarily, the training dataset suffers from a lack of diversity and high-quality samples, as well as cross-paired data, i.e., paired samples whose components originate from different instances. In addition, the current mechanism for integrating multiple reference images is suboptimal, potentially resulting in the confusion of multiple subjects. To overcome these limitations, we propose a dedicated data construction pipeline, incorporating low-quality sample filtering and diverse data synthesis, to produce consistency-preserving training data. Moreover, we introduce Reference Rotary Positional Encoding (R-RoPE) to process reference images, enabling stable and precise multi-image integration. Extensive experiments across numerous benchmarks demonstrate that Kaleido significantly outperforms previous methods in consistency, fidelity, and generalization, marking an advance in S2V generation.
Similar Papers
Scaling Sequence-to-Sequence Generative Neural Rendering
CV and Pattern Recognition
Creates realistic 3D scenes from pictures.
Scaling Zero-Shot Reference-to-Video Generation
CV and Pattern Recognition
Makes videos match text and pictures.
MultiRef: Controllable Image Generation with Multiple Visual References
CV and Pattern Recognition
Lets computers create art from many pictures.