[论文] [论文] FL4N: Feed-forward Latent Model for 4D Novel View Synthesis

论文概要

研究领域: CV
作者: Eunbeen Hong, Hyeonseo Yu, Honggyu An, Jooho Son, Hyunah Ko et al.
发布时间: 2026-10-26
arXiv: 2610.35854

中文摘要

4D 新视角合成（NVS）旨在跨视点与跨时刻渲染动态场景，需要联合推断 3D 结构与时间动态。现有单目 feed-forward 方法多依赖显式 4D 高斯表示，容易产生时间不一致与严重渲染伪影。我们提出 FL4N——首个 feed-forward 隐空间 4D NVS 模型。它不依赖显式 4D 基元，而是将 4D 场景先验编码到统一的时空隐表示中。核心洞察：在预训练静态隐先验之上把动态学成残差，就能在动态训练数据有限时有效完成 4D 建模。为此引入轻量动态编码器，利用静态场景特征将物体运动与相机视差解耦。仅以图像合成目标训练，FL4N 达到 SOTA 的 4D NVS 质量并实时渲染；学到的隐表示可泛化到真实视频，并把视点与时间解耦。

原文摘要

4D novel view synthesis (NVS) aims to render dynamic scenes across viewpoints and timestamps, which requires jointly inferring 3D structure and temporal dynamics. Existing monocular feed-forward methods often rely on explicit 4D Gaussian representations, which are prone to temporal inconsistency and severe rendering artifacts. We present FL4N, the first feed-forward latent 4D NVS model. Instead of relying on explicit 4D primitives, FL4N encodes a 4D scene prior into a unified latent representation across space and time. Our core insight is that learning temporal dynamics as residuals on top of a pretrained static latent prior enables effective 4D modeling with limited dynamic training data. To this end, we introduce a lightweight dynamic encoder that leverages static scene features to disentangle object motion from camera-induced parallax. Trained solely with an image synthesis objective, FL4N achieves state-of-the-art 4D NVS quality while rendering in real time. Its learned latent representation generalizes to real-world videos and disentangles viewpoint from time.

自动采集于 2026-10-28

#论文 #arXiv #CV #小凯