Pose-Free Generalizable Rendering Transformer

TL;DR

PF-GRT introduces a pose-free rendering transformer that learns to synthesize novel views without pre-computed camera poses, achieving high-quality, generalizable, and robust view synthesis across diverse datasets.

Contribution

The paper presents PF-GRT, a novel framework that eliminates the need for camera pose estimation in view synthesis by learning feature matching directly from data.

Findings

Achieves superior photo-realistic image quality in zero-shot rendering tasks.

Demonstrates robustness to noisy camera pose inputs.

Generalizes well to unseen scenes without pose information.

Abstract

In the field of novel-view synthesis, the necessity of knowing camera poses (e.g., via Structure from Motion) before rendering has been a common practice. However, the consistent acquisition of accurate camera poses remains elusive, and errors in pose extraction can adversely impact the view synthesis process. To address this challenge, we introduce PF-GRT, a new Pose-Free framework for Generalizable Rendering Transformer, eliminating the need for pre-computed camera poses and instead leveraging feature-matching learned directly from data. PF-GRT is parameterized using a local relative coordinate system, where one of the source images is set as the origin. An OmniView Transformer is designed for fusing multi-view cues under the pose-free setting, where unposed-view fusion and origin-centric aggregation are performed. The 3D point feature along target ray is sampled by projecting onto the selected origin plane. The final pixel intensities are modulated and decoded using another Transformer. PF-GRT demonstrates an impressive ability to generalize to new scenes that were not encountered during the training phase, without the need of pre-computing camera poses. Our experiments with zero-shot rendering on the LLFF, RealEstate-10k, Shiny, and Blender datasets reveal that it produces superior quality in generating photo-realistic images. Moreover, it demonstrates robustness against noise in test camera poses. Code is available at https://zhiwenfan.github.io/PF-GRT/.