GeoBench: Benchmarking and Analyzing Monocular Geometry Estimation Models

TL;DR

This paper introduces GeoBench, a comprehensive benchmark for monocular geometry estimation, revealing that discriminative models with high-quality synthetic data can outperform generative models, emphasizing data quality over scale.

Contribution

It provides a unified evaluation framework and new challenging benchmarks, highlighting the importance of data quality in geometry estimation models.

Findings

Discriminative models like DINOv2 outperform generative models with synthetic data.

High-quality synthetic data can lead to state-of-the-art results with simpler models.

Benchmarking on diverse scenes reveals the impact of data quality on performance.

Abstract

Recent advances in discriminative and generative pretraining have yielded geometry estimation models with strong generalization capabilities. While discriminative monocular geometry estimation methods rely on large-scale fine-tuning data to achieve zero-shot generalization, several generative-based paradigms show the potential of achieving impressive generalization performance on unseen scenes by leveraging pre-trained diffusion models and fine-tuning on even a small scale of synthetic training data. Frustratingly, these models are trained with different recipes on different datasets, making it hard to find out the critical factors that determine the evaluation performance. Besides, current geometry evaluation benchmarks have two main drawbacks that may prevent the development of the field, i.e., limited scene diversity and unfavorable label quality. To resolve the above issues, (1) we build fair and strong baselines in a unified codebase for evaluating and analyzing the geometry estimation models; (2) we evaluate monocular geometry estimators on more challenging benchmarks for geometry estimation task with diverse scenes and high-quality annotations. Our results reveal that pre-trained using large data, discriminative models such as DINOv2, can outperform generative counterparts with a small amount of high-quality synthetic data under the same training configuration, which suggests that fine-tuning data quality is a more important factor than the data scale and model architecture. Our observation also raises a question: if simply fine-tuning a general vision model such as DINOv2 using a small amount of synthetic depth data produces SOTA results, do we really need complex generative models for depth estimation? We believe this work can propel advancements in geometry estimation tasks as well as a wide range of downstream applications.