Zero-Shot Scene Reconstruction from Single Images with Deep Prior Assembly

TL;DR

This paper introduces deep prior assembly, a zero-shot framework that combines large model priors for accurate scene reconstruction from single images without additional training.

Contribution

It proposes a novel method to assemble diverse deep priors for scene reconstruction, eliminating the need for 3D or 2D data-driven training.

Findings

Demonstrates superior performance on various datasets.

Achieves robust scene reconstruction without extra data.

Outperforms recent methods in visual and numerical evaluations.

Abstract

Large language and vision models have been leading a revolution in visual computing. By greatly scaling up sizes of data and model parameters, the large models learn deep priors which lead to remarkable performance in various tasks. In this work, we present deep prior assembly, a novel framework that assembles diverse deep priors from large models for scene reconstruction from single images in a zero-shot manner. We show that this challenging task can be done without extra knowledge but just simply generalizing one deep prior in one sub-task. To this end, we introduce novel methods related to poses, scales, and occlusion parsing which are keys to enable deep priors to work together in a robust way. Deep prior assembly does not require any 3D or 2D data-driven training in the task and demonstrates superior performance in generalizing priors to open-world scenes. We conduct evaluations on various datasets, and report analysis, numerical and visual comparisons with the latest methods to show our superiority. Project page: https://junshengzhou.github.io/DeepPriorAssembly.