TR2M: Transferring Monocular Relative Depth to Metric Depth with Language Descriptions and Dual-Level Scale-Oriented Contrast

TL;DR

TR2M introduces a novel framework that leverages language descriptions and dual-level contrastive learning to accurately transfer relative depth to metric depth across diverse domains, enhancing generalization and zero-shot performance.

Contribution

The paper proposes TR2M, a new method combining text and image inputs with scale-oriented contrastive learning to improve metric depth estimation from relative depth data.

Findings

TR2M achieves state-of-the-art results on multiple datasets.

The method demonstrates strong zero-shot generalization.

It effectively reduces scale uncertainty in depth estimation.

Abstract

This work presents a generalizable framework to transfer relative depth to metric depth. Current monocular depth estimation methods are mainly divided into metric depth estimation (MMDE) and relative depth estimation (MRDE). MMDEs estimate depth in metric scale but are often limited to a specific domain. MRDEs generalize well across different domains, but with uncertain scales which hinders downstream applications. To this end, we aim to build up a framework to solve scale uncertainty and transfer relative depth to metric depth. Previous methods used language as input and estimated two factors for conducting rescaling. Our approach, TR2M, utilizes both text description and image as inputs and estimates two rescale maps to transfer relative depth to metric depth at pixel level. Features from two modalities are fused with a cross-modality attention module to better capture scale information. A strategy is designed to construct and filter confident pseudo metric depth for more comprehensive supervision. We also develop scale-oriented contrastive learning to utilize depth distribution as guidance to enforce the model learning about intrinsic knowledge aligning with the scale distribution. TR2M only exploits a small number of trainable parameters to train on datasets in various domains and experiments not only demonstrate TR2M's great performance in seen datasets but also reveal superior zero-shot capabilities on five unseen datasets. We show the huge potential in pixel-wise transferring relative depth to metric depth with language assistance. (Code is available at: https://github.com/BeileiCui/TR2M)