This paper presents MonoRelief V2, an end-to-end model designed for directly recovering 2.5D reliefs from single images under complex material and illumination variations. In contrast to its predecessor, MonoRelief V1 [1], which was solely trained on synthetic data, MonoRelief V2 incorporates real data to achieve improved robustness, accuracy and efficiency. To overcome the challenge of acquiring large-scale real-world dataset, we generate approximately 15,000 pseudo real images using a text-to-image generative model, and derive corresponding depth pseudo-labels through fusion of depth and normal predictions. Furthermore, we construct a small-scale real-world dataset (800 samples) via multi-view reconstruction and detail refinement. MonoRelief V2 is then progressively trained on the pseudo-real and real-world datasets. Comprehensive experiments demonstrate its state-of-the-art performance both in depth and normal predictions, highlighting its strong potential for a range of downstream applications. Code is at: https://github.com/glp1001/MonoreliefV2.
@article{arxiv.2508.19555,
title = {MonoRelief V2: Leveraging Real Data for High-Fidelity Monocular Relief Recovery},
author = {Yu-Wei Zhang and Tongju Han and Lipeng Gao and Mingqiang Wei and Hui Liu and Changbao Li and Caiming Zhang},
journal= {arXiv preprint arXiv:2508.19555},
year = {2025}
}