Research map / 3D vision / Depth and stereo
SeeClear: Reliable Transparent Object Depth Estimation via Generative Opacification
Xiaoying Wang, Yumeng He, Jingkai Shi, Jiayin Lu, Yin Yang, Ying Jiang, Chenfanfu Jiang
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 |
|---|---|---|---|
| add generative opacification front-end to Depth Anything V3 | DA3 · ClearGrasp Real (ToM region) | RMSE (mm): 24.03 -> 21.45 | Improved |
| add generative opacification front-end to MoGe-2 | MoGe-2 · ClearGrasp Real (ToM region) | RMSE (mm): 56.83 -> 26.62 | Improved |
| add generative opacification front-end to MoGe-2 | MoGe-2 · TransPhy3D (synthetic, full image) | AbsRel: 0.016 -> 0.012 | Improved |
| add generative opacification front-end to DA3 | DA3 · ClearGrasp Real (All region) | RMSE (mm): 38.88 -> 34.85 | Improved |
| use segmentation mask instead of bounding box for localization (ablation) | SeeClear · ClearGrasp Real (ToM region) (DA3 backbone) | RMSE (mm): 22.96 -> 21.45 | Improved |
| use CLIP class token instead of DINOv2 CLS token (ablation) | SeeClear · ClearGrasp Real (ToM region) (DA3 backbone) | RMSE (mm): 23.85 -> 21.45 | Improved |
| add alpha blending in compositing (ablation) | SeeClear · ClearGrasp Real (ToM region) (DA3 backbone) | RMSE (mm): 23.60 -> 21.45 | Improved |
| add mask refinement module (ablation) | SeeClear · ClearGrasp Real (ToM region) (DA3 backbone) | RMSE (mm): 21.73 -> 21.45 | No clear effect |
| use LPIPS loss instead of L1 loss (ablation) | SeeClear · ClearGrasp Real (ToM region) (DA3 backbone) | MAE (mm): 18.19 -> 17.94 | Improved |
| use LPIPS loss instead of gradient loss (ablation) | SeeClear · ClearGrasp Real (ToM region) (DA3 backbone) | RMSE (mm): 22.21 -> 21.45 | Improved |
| use generative opacification instead of solid-color fill (ablation) | SeeClear · ClearGrasp Real (ToM region) (DA3 backbone) | RMSE (mm): 33.09 -> 21.45 | Improved |
| compare zero-shot SeeClear (DA3) with fine-tuned DKT | SeeClear (DA3) · ClearGrasp Real (ToM region) | delta<1.025 (%): 51.27 vs 52.69 | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MoGe-2 — MoGe-2: Accurate Monocular Geometry with Metric Scale and Sharp Details (NeurIPS 2025) · better on 4 of 4 shared results
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
- Seeing and Seeing Through the Glass: Real and Synthetic Data for Multi-Layer Depth Estimation (ICCV 2025 · similarity 0.88)
- SeeGroup: Multi-Layer Depth Estimation of Transparent Surfaces via Self-Determined Grouping (CVPR 2026 · similarity 0.88)
- PolarDepth: Monocular Transparent Object Depth from Polar-Physics Priors (ICML 2026 · similarity 0.87)
- From Transparent to Opaque: Rethinking Neural Implicit Surfaces with $\alpha$-NeuS (NeurIPS 2024 · similarity 0.86)
- DepthFM: Fast Generative Monocular Depth Estimation with Flow Matching (AAAI 2025 · similarity 0.86)
- Pixel-Perfect Depth with Semantics-Prompted Diffusion Transformers (NeurIPS 2025 · similarity 0.85)
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.
