The year/Independent research

Paper 2603.18524

3DreamBooth: High-Fidelity 3D Subject-Driven Video Generation Model

Published
Mar 2026
Research lab
Independent
Citations
0
GitHub
60 stars

01 In brief

Summary

The paper introduces 3DreamBooth, a framework for 3D-aware video customization that generates view-consistent videos of a subject from a few multi-view reference images.

It addresses the limitation of existing subject-driven video generation methods that treat subjects as 2D entities, lacking 3D geometry priors.

The framework comprises two components: 3DreamBooth, which uses a 1-frame optimization paradigm to decouple spatial geometry from temporal motion, baking a 3D prior into the model without video-based training; and 3Dapter, a multi-view conditioning module that injects fine-grained textures and accelerates convergence.

3Dapter is pre-trained on single-view data and then jointly optimized with 3DreamBooth using an asymmetrical conditioning strategy, acting as a dynamic selective router that queries view-specific geometric hints.

The authors also introduce 3D-CustomBench, a benchmark for evaluating 3D-consistent video customization.

Experiments show the framework outperforms baselines (VACE, Phantom) in multi-view subject fidelity, 3D geometric fidelity (Chamfer Distance), and video quality, while maintaining computational efficiency.

02 From the paper

Abstract

Creating dynamic, view-consistent videos of customized subjects is highly sought after for a wide range of emerging applications, including immersive VR/AR, virtual production, and next-generation e-commerce. However, despite rapid progress in subject-driven video generation, existing methods predominantly treat subjects as 2D entities, focusing on transferring identity through single-view visual features or textual prompts. Because real-world subjects are inherently 3D, applying these 2D-centric approaches to 3D object customization reveals a fundamental limitation: they lack the comprehensive spatial priors necessary to reconstruct the 3D geometry. Consequently, when synthesizing novel views, they must rely on generating plausible but arbitrary details for unseen regions, rather than preserving the true 3D identity. Achieving genuine 3D-aware customization remains challenging due to the scarcity of multi-view video datasets. While one might attempt to fine-tune models on limited video sequences, this often leads to temporal overfitting. To resolve these issues, we introduce a novel framework for 3D-aware video customization, comprising 3DreamBooth and 3Dapter. 3DreamBooth decouples spatial geometry from temporal motion through a 1-frame optimization paradigm. By restricting updates to spatial representations, it effectively bakes a robust 3D prior into the model without the need for exhaustive video-based training. To enhance fine-grained textures and accelerate convergence, we incorporate 3Dapter, a visual conditioning module. Following single-view pre-training, 3Dapter undergoes multi-view joint optimization with the main generation branch via an asymmetrical conditioning strategy. This design allows the module to act as a dynamic selective router, querying view-specific geometric hints from a minimal reference set. Project page: https://ko-lani.github.io/3DreamBooth/