DiVA enables smooth long-term interactive digital character videos
DiVA: Enabling Interactive Digital Life Simulation via Video Models
Computer Vision and Pattern Recognition
Summary
Making digital characters interact naturally over long periods is hard because video quality and motion can degrade. The authors created DiVA, a system that carefully manages video generation in three parts, including special transitions that keep the characters stable and realistic. This approach fixes common problems like shaky cameras and awkward pose changes, making conversations with video avatars look smoother and more lifelike. DiVA was tested against existing methods and showed better video quality and continuity during extended interactions.
What this means in practice
- •For game developers: Create interactive digital characters with smooth video responses that maintain visual quality over long gameplay sessions.
- •For virtual assistant developers: Build video-based avatars that can engage in multi-turn conversations with realistic actions and stable visual continuity.
Authors
Cheng Chen, Hao Ouyang, Qiuyu Wang, Ka Leong Cheng, Wen Wang, Yihao Meng, Hanlin Wang, Yixuan Li, Jiacheng Wei, Zhenshan Tan, Yanhong Zeng, Yujun Shen, Guosheng Lin, Fayao Liu
Abstract
We present DiVA, a deeply interactive digital life simulator pioneering a new paradigm for long-term, open-ended interactive experiences within digital character worlds. DiVA's architecture pairs a Multimodal Large Language Model (MLLM) as a router with a meticulously designed stacked video pipeline for seamless, multi-turn interactions with action and audio response. To maintain continuity and avoid degradation, we model generation as a three-part coupled system: waiting video, action video, and the transitions between them. These transitions are critically handled by our Anchored Video Continuation (AVC) module, which returns the character to stable states to prevent degradation. By encoding information from the preceding action video segment, AVC ensures smooth transitions, significantly reducing camera jitter and inconsistencies common in current video transition methods. This design also enables complex pose changes (e.g., sitting to standing) typically difficult for audio-driven models. These system designs together ensure high-fidelity identity, coherence, and dynamics for extended experiences. To validate our pipeline design, we comprehensively compare our system against alternatives by replacing our core generation module with mainstream long-video, continuation, and interpolation methods. We further analyze the necessity of the three-stage design, anchor-state selection, transition naturalness, spatial grounding, and the quality-latency trade-off, and we expand the comparison to additional long-form audio-driven avatar models. Results confirm DiVA is markedly superior in maintaining long-term visual quality and realism, validating its effectiveness as a sustainable, interactive simulation.