Research map / 3D vision / Depth and stereo
DiffusionDepth: Diffusion Denoising Approach for Monocular Depth Estimation
Yiqun Duan, Xianda Guo, Zheng Zhu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| reformulate depth as diffusion denoising vs best prior (URCDC-Depth) on official KITTI split | DiffusionDepth · KITTI official offline split | RMSE: 1.528 -> 1.418 | Improved |
| diffusion denoising head vs VA-Depth on NYU-Depth-v2 | DiffusionDepth · NYU-Depth-v2 | RMSE: 0.304 -> 0.295 | Improved |
| diffusion head vs VA-Depth/URCDC-Depth on KITTI online server | DiffusionDepth · KITTI online server | SILog: 9.84 vs 9.85 | No clear effect |
| diffuse refined depth latent (self-diffusion) instead of sparse GT (ablation) | DiffusionDepth · KITTI official split | RMSE: 12.3772 -> 1.4523 | Improved |
| diffuse refined depth instead of dense GT on indoor data (ablation) | DiffusionDepth · NYU-Depth-V2 | RMSE: 0.3041 -> 0.2983 | No clear effect |
| reduce inference steps from 20 to 5 without retraining (ablation) | DiffusionDepth · NYU-Depth-V2 | RMSE: 0.298 -> 1.1750 | Hurt |
| train with fewer inference steps (5) instead of 20 (ablation) | DiffusionDepth · NYU-Depth-V2 | RMSE: 0.298 -> 0.4366 | Hurt |
| use HAHI neck with Swin instead of FPN (ablation) | DiffusionDepth · KITTI official split | RMSE: 1.5569 -> 1.4523 | Improved |
| use Swin transformer backbone instead of ResNet-50 (ablation) | DiffusionDepth · KITTI official split | RMSE: 1.7124 -> 1.5569 | Improved |
| use depth latent at x2 downsampling instead of x4 (ablation) | DiffusionDepth · KITTI official split | RMSE: 1.508 -> 1.4523 | Improved |
Later papers that compared with this method
1 later accepted paper reports DiffusionDepth as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Hyper-Depth: Hypergraph-based Multi-Scale Representation Fusion for Monocular Depth Estimation (ICCV 2025 · better on 1 of 1 shared result)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Repurposing Diffusion-Based Image Generators for Monocular Depth Estimation (CVPR 2024 · similarity 0.90)
- Pixel-Perfect Depth with Semantics-Prompted Diffusion Transformers (NeurIPS 2025 · similarity 0.89)
- DepthFM: Fast Generative Monocular Depth Estimation with Flow Matching (AAAI 2025 · similarity 0.89)
- BetterDepth: Plug-and-Play Diffusion Refiner for Zero-Shot Monocular Depth Estimation (NeurIPS 2024 · similarity 0.88)
- SharpDepth: Sharpening Metric Depth Predictions Using Diffusion Distillation (CVPR 2025 · similarity 0.88)
- Jasmine: Harnessing Diffusion Prior for Self-supervised Depth Estimation (NeurIPS 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
