3D-VirtFusion: Synthetic 3D Data Augmentation through Generative Diffusion Models and Controllable Editing

TL;DR

This paper introduces 3D-VirtFusion, a novel approach that leverages diffusion models and large language models to automatically generate diverse 3D training data, improving scene understanding and addressing data scarcity issues.

Contribution

It presents a new method for synthetic 3D data augmentation using pretrained foundation models, enabling high-level diversity and controllable editing of 3D scenes.

Findings

Generated diverse 3D scene data without real data

Enhanced model performance in few-shot learning scenarios

Mitigated class imbalance issues in 3D datasets

Abstract

Data augmentation plays a crucial role in deep learning, enhancing the generalization and robustness of learning-based models. Standard approaches involve simple transformations like rotations and flips for generating extra data. However, these augmentations are limited by their initial dataset, lacking high-level diversity. Recently, large models such as language models and diffusion models have shown exceptional capabilities in perception and content generation. In this work, we propose a new paradigm to automatically generate 3D labeled training data by harnessing the power of pretrained large foundation models. For each target semantic class, we first generate 2D images of a single object in various structure and appearance via diffusion models and chatGPT generated text prompts. Beyond texture augmentation, we propose a method to automatically alter the shape of objects within 2D images. Subsequently, we transform these augmented images into 3D objects and construct virtual scenes by random composition. This method can automatically produce a substantial amount of 3D scene data without the need of real data, providing significant benefits in addressing few-shot learning challenges and mitigating long-tailed class imbalances. By providing a flexible augmentation approach, our work contributes to enhancing 3D data diversity and advancing model capabilities in scene understanding tasks.