ConsID-Gen：视角一致且身份保持的图像到视频生成
ConsID-Gen: View-Consistent and Identity-Preserving Image-to-Video Generation
用无位姿辅助视角增强首帧、经双流视觉-几何编码器与文本-视觉连接器融合语义与结构线索，为 Diffusion Transformer 提供统一条件，可直接借鉴到任何需要多视角一致性的条件视频生成任务中。
原文依据 augments the first frame with unposed auxiliary views and fuses semantic and structural cues via a dual-stream visual-geometric encoder
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论图像到视频生成（I2V）按照文本指令将静态图像动画化为时序连贯的视频序列，但在视角变化下保持细粒度物体身份仍是一个持续存在的挑战。与文本到视频模型不同，现有 I2V 流程常出现外观漂移与几何畸变，我们将这些伪影归因于单视角 2D 观测的稀疏性以及跨模态对齐的不足。本文从数据与模型两个层面解决该问题。首先，我们构建了 ConsIDVid，这是一个以物体为中心的大规模数据集，通过可扩展的流程生成高质量、时序对齐的视频；并建立了 ConsIDVid-Bench，提出一种新的多视角一致性基准与评测框架，其度量对细微的几何与外观偏差敏感。我们进一步提出 ConsID-Gen，一种视角辅助的 I2V 生成框架，用无位姿的辅助视角增强首帧，并通过双流视觉-几何编码器以及文本-视觉连接器融合语义与结构线索，为 Diffusion Transformer 主干提供统一条件。在 ConsIDVid-Bench 上的实验表明，ConsID-Gen 在多项指标上持续领先，最佳整体性能超过 Wan2.1 与 HunyuanVideo 等领先视频生成模型，在具有挑战性的真实场景下实现更优的身份保真度与时序连贯性。我们的模型与数据集见 https://myangwu.github.io/ConsID-Gen。
英文原文摘要 1548 字符
Image-to-Video generation (I2V) animates a static image into a temporally coherent video sequence following textual instructions, yet preserving fine-grained object identity under changing viewpoints remains a persistent challenge. Unlike text-to-video models, existing I2V pipelines often suffer from appearance drift and geometric distortion, artifacts we attribute to the sparsity of single-view 2D observations and weak cross-modal alignment. Here we address this problem from both data and model perspectives. First, we curate ConsIDVid, a large-scale object-centric dataset built with a scalable pipeline for high-quality, temporally aligned videos, and establish ConsIDVid-Bench, where we present a novel benchmarking and evaluation framework for multi-view consistency using metrics sensitive to subtle geometric and appearance deviations. We further propose ConsID-Gen, a view-assisted I2V generation framework that augments the first frame with unposed auxiliary views and fuses semantic and structural cues via a dual-stream visual-geometric encoder as well as a text-visual connector, yielding unified conditioning for a Diffusion Transformer backbone. Experiments across ConsIDVid-Bench demonstrate that ConsID-Gen consistently outperforms in multiple metrics, with the best overall performance surpassing leading video generation models like Wan2.1 and HunyuanVideo, delivering superior identity fidelity and temporal coherence under challenging real-world scenarios. Our model and dataset are at https://myangwu.github.io/ConsID-Gen.