Research map / 3D vision / Depth and stereo
Atlantis++: Enabling Underwater Depth Estimation with Stable Diffusion and Beyond
Fan Zhang, Shaodi You, Yu Li, Ying Fu
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 |
|---|---|---|---|
| train on Atlantis instead of NYUDepthv2 pretraining | iDisc · Sea-thru D3+D5 | RMSE: 3.144 -> 1.371 | Improved |
| train on Atlantis instead of NYUDepthv2 pretraining | iDisc · Sea-thru D3+D5 | delta1: 0.215 -> 0.553 | Improved |
| train on Atlantis instead of KITTI pretraining | NeWCRFs · Sea-thru D3+D5 | RMSE: 3.251 -> 1.435 | Improved |
| train on Atlantis instead of KITTI pretraining | NeWCRFs · Sea-thru D3+D5 | delta1: 0.213 -> 0.476 | Improved |
| train on Atlantis instead of KITTI pretraining | iDisc · SQUID | RMSE: 7.265 -> 2.663 | Improved |
| train on Atlantis instead of KITTI pretraining | iDisc · SQUID | delta1: 0.156 -> 0.637 | Improved |
| train on Atlantis instead of KITTI pretraining | NeWCRFs · SQUID | RMSE: 6.692 -> 2.563 | Improved |
| train on Atlantis instead of KITTI pretraining | NeWCRFs · SQUID | delta1: 0.197 -> 0.675 | Improved |
| train on Atlantis instead of KITTI pretraining | NeWCRFs · Sea-thru D3+D5 | S.Rel: 15.768 -> 14.764 | No clear effect |
| train on Atlantis instead of NYUDepthv2 pretraining | NeWCRFs · Sea-thru D3+D5 | A.Rel: 0.770 -> 1.683 | Mixed |
| generate data with Stable Diffusion+ControlNet instead of GAN style transfer | synthetic underwater dataset · LLaVA v1.5 underwater recognition | recognized as underwater | Improved |
Closest papers by meaning
- Atlantis: Enabling Underwater Depth Estimation with Stable Diffusion (CVPR 2024 · similarity 0.99)
- Sea-ing in Low-light (CVPR 2025 · similarity 0.88)
- NAUTILUS: A Large Multimodal Model for Underwater Scene Understanding (NeurIPS 2025 · similarity 0.88)
- DepthFM: Fast Generative Monocular Depth Estimation with Flow Matching (AAAI 2025 · similarity 0.87)
- WaterDiffusion: Learning a Prior-involved Unrolling Diffusion for Joint Underwater Saliency Detection and Visual Restoration (AAAI 2025 · similarity 0.87)
- Plenodium: Underwater 3D Scene Reconstruction with Plenoptic Medium Representation (NeurIPS 2025 · similarity 0.87)
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.
