CVPR Findings2026
Re-Depth Anything: Test-Time Depth Refinement via Self-Supervised Re-lighting
Ananta R. Bhattarai, Helge Rhodin
Abstract
Input Image DA-V2 Ours Frontal Left Novel Views Right Novel Views Test-time Similar Real Tiger Image Figure 1. Re-Depth Anything refines the predictions of Depth Anything V2 [53] by re-lighting the reconstructed geometry and extracting knowledge from diffusion models in a self-supervised manner. In this example, test-time optimization enhances facial detail (see frontal view) and refines the nose shape to look more like a tiger (side view), correcting the dog-like initial resemblance likely originating from a biased training distribution. The key contribution is a re-synthesis method that replaces photometric reconstruction for self-supervision.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
Builds on34
- An Image is Worth 16x16 Words: Transformers for Image Recognition at ScaleAlexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn et al.ICLR 2021 · 21,477 citations
- High-Resolution Image Synthesis with Latent Diffusion ModelsRobin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser et al.CVPR 2022 · 13,123 citations
- Photorealistic Text-to-Image Diffusion Models with Deep Language UnderstandingChitwan Saharia, William Chan, Saurabh Saxena, Lala Li et al.NeurIPS 2022 · 8,965 citations
- BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language ModelsJunnan Li, Dongxu Li, Silvio Savarese, Steven C. H. HoiICML 2023 · 7,873 citations
- 3D Gaussian Splatting for Real-Time Radiance Field RenderingBernhard Kerbl, Georgios Kopanas, Thomas Leimkühler, George DrettakisSIGGRAPH 2023 · 5,687 citations
Related papers
- Depth Anything V2Lihe Yang, Bingyi Kang, Zilong Huang, Zhen Zhao et al.NeurIPS 2024 · 2,305 citations
- Test3R: Learning to Reconstruct 3D at Test TimeYuheng Yuan, Qiuhong Shen, Shizun Wang, Xingyi Yang et al.NeurIPS 2025 · 19 citations
- SfM-TTR: Using Structure from Motion for Test-Time Refinement of Single-View Depth NetworksSergio Izquierdo, Javier CiveraCVPR 2023
- Towards High Fidelity Monocular Face Reconstruction with Rich Reflectance using Self-supervised Learning and Ray TracingAbdallah Dib, Cédric Thébault, Junghyun Ahn, Philippe-Henri Gosselin et al.ICCV 2021 · 63 citations
- Differentiable Diffusion for Dense Depth Estimation From Multi-View ImagesNumair Khan, Min H. Kim, James TompkinCVPR 2021
