Anchored Video Generation: Decoupling Scene Construction and Temporal Synthesis in Text-to-Video Diffusion Models

TL;DR

This paper introduces Anchored Video Generation, a modular approach that improves text-to-video synthesis by decoupling scene construction and temporal animation, leading to better scene consistency, efficiency, and control.

Contribution

The paper proposes a novel three-stage pipeline that separates reasoning, composition, and temporal synthesis, significantly enhancing video quality and efficiency in text-to-video models.

Findings

Sets new state-of-the-art on T2V CompBench benchmark

Improves all tested models on VBench2

Reduces sampling steps by 70% without performance loss

Abstract

State-of-the-art Text-to-Video (T2V) diffusion models can generate visually impressive results, yet they still frequently fail to compose complex scenes or follow logical temporal instructions. In this paper, we argue that many errors, including apparent motion failures, originate from the model's inability to construct a semantically correct or logically consistent initial frame. We introduce Anchored Video Generation (AVG), a modular pipeline that decouples these tasks by decomposing the Text-to-Video generation into three specialized stages: (1) Reasoning, where a Large Language Model (LLM) rewrites the video prompt to describe only the initial scene, resolving temporal ambiguities; (2) Composition, where a Text-to-Image (T2I) model synthesizes a high-quality, compositionally-correct anchor frame from this new prompt; and (3) Temporal Synthesis, where a video model, finetuned to understand this anchor, focuses its entire capacity on animating the scene and following the prompt. Our approach sets a new state-of-the-art on the T2V CompBench benchmark and significantly improves all tested models on VBench2. Furthermore, we show that visual anchoring allows us to cut the number of sampling steps by 70% without any loss in performance. AVG offers a simple yet practical path toward more efficient, robust, and controllable video synthesis.