MaRINeR: Enhancing Novel Views by Matching Rendered Images with Nearby References

TL;DR

MaRINeR is a refinement technique that enhances novel view renderings by matching and transferring details from nearby reference images, improving quality in 3D scene reconstructions.

Contribution

It introduces a novel reference-based refinement method that leverages deep feature matching to improve rendering quality from imperfect 3D reconstructions.

Findings

Improved rendering quality in quantitative metrics.

Enhanced detail recovery in low-quality reconstructions.

Effective in downstream tasks like data augmentation and validation.

Abstract

Rendering realistic images from 3D reconstruction is an essential task of many Computer Vision and Robotics pipelines, notably for mixed-reality applications as well as training autonomous agents in simulated environments. However, the quality of novel views heavily depends of the source reconstruction which is often imperfect due to noisy or missing geometry and appearance. Inspired by the recent success of reference-based super-resolution networks, we propose MaRINeR, a refinement method that leverages information of a nearby mapping image to improve the rendering of a target viewpoint. We first establish matches between the raw rendered image of the scene geometry from the target viewpoint and the nearby reference based on deep features, followed by hierarchical detail transfer. We show improved renderings in quantitative metrics and qualitative examples from both explicit and implicit scene representations. We further employ our method on the downstream tasks of pseudo-ground-truth validation, synthetic data enhancement and detail recovery for renderings of reduced 3D reconstructions.