InfiniDepth: Arbitrary-Resolution and Fine-Grained Depth Estimation with Neural Implicit Fields

TL;DR

InfiniDepth introduces a neural implicit field approach for depth estimation, enabling arbitrary resolution and fine detail recovery, outperforming existing methods on synthetic and real benchmarks, and enhancing view synthesis quality.

Contribution

The paper presents a novel neural implicit depth representation allowing continuous, high-resolution depth estimation and detailed geometric recovery, surpassing traditional grid-based methods.

Findings

Achieves state-of-the-art results on synthetic and real benchmarks.

Excels in fine-detail regions and large viewpoint shifts.

Produces higher quality novel view synthesis with fewer artifacts.

Abstract

Existing depth estimation methods are fundamentally limited to predicting depth on discrete image grids. Such representations restrict their scalability to arbitrary output resolutions and hinder the geometric detail recovery. This paper introduces InfiniDepth, which represents depth as neural implicit fields. Through a simple yet effective local implicit decoder, we can query depth at continuous 2D coordinates, enabling arbitrary-resolution and fine-grained depth estimation. To better assess our method's capabilities, we curate a high-quality 4K synthetic benchmark from five different games, spanning diverse scenes with rich geometric and appearance details. Extensive experiments demonstrate that InfiniDepth achieves state-of-the-art performance on both synthetic and real-world benchmarks across relative and metric depth estimation tasks, particularly excelling in fine-detail regions. It also benefits the task of novel view synthesis under large viewpoint shifts, producing high-quality results with fewer holes and artifacts.