Research map / 3D vision / Depth and stereo
Deep Reinforcement Learning of Volume-Guided Progressive View Inpainting for 3D Point Scene Completion From a Single Depth Image
Xiaoguang Han, Zhaoxuan Zhang, Dong Du, Mingdai Yang, Jingming Yu, Pan Pan, Xin Yang, Ligang Liu, Zixiang Xiong, Shuguang Cui
Check 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 volume guidance to depth inpainting (ablation) | Volume-guided view inpainting · SUNCG depth inpainting | L^1_Ω: 0.0717 -> 0.0470 | Improved |
| Add volume guidance to depth inpainting (ablation) | Volume-guided view inpainting · SUNCG depth inpainting | PSNR: 22.15 -> 24.73 | Improved |
| Add volume guidance to depth inpainting (ablation) | Volume-guided view inpainting · SUNCG depth inpainting | SSIM: 0.910 -> 0.930 | Improved |
| Add projection back-propagation from the 2D inpainting network to the 3D volume network (ablation) | Volume-guided view inpainting · SUNCG depth inpainting | L^1_Ω: 0.0574 -> 0.0470 | Improved |
| Add projection back-propagation from the 2D inpainting network to the 3D volume network (ablation) | Volume-guided view inpainting · SUNCG depth inpainting | PSNR: 23.12 -> 24.73 | Improved |
| Add projection back-propagation from the 2D inpainting network to the 3D volume network (ablation) | Volume-guided view inpainting · SUNCG depth inpainting | SSIM: 0.926 -> 0.930 | Improved |
| Add the hole-area reward to DQN view planning (ablation) | DQN progressive view planning · SUNCG scene completion | CD: 0.1495 -> 0.1148 | Improved |
| Replace DQN view selection with 10 uniformly sampled views (ablation) | Progressive scene completion · SUNCG scene completion | CD: 0.1148 -> 0.1841 | Hurt |
| Replace DQN view selection with 5 uniformly sampled views (ablation) | Progressive scene completion · SUNCG scene completion | CD: 0.1148 -> 0.1642 | Hurt |
| Use 10 rather than 5 uniformly sampled views (ablation) | Uniform-view progressive scene completion · SUNCG scene completion | CD: 0.1642 -> 0.1841 | Hurt |
Closest papers by meaning
- Point Cloud Scene Completion With Joint Color and Semantic Estimation From Single RGB-D Image (TPAMI 2023 · similarity 0.93)
- Semantic Scene Completion From a Single Depth Image (CVPR 2017 · similarity 0.87)
- Contextual-based Image Inpainting: Infer, Match, and Translate (ECCV 2018 · similarity 0.87)
- VolumeFusion: Deep Depth Fusion for 3D Scene Reconstruction (ICCV 2021 · similarity 0.86)
- InstaInpaint: Instant 3D-Scene Inpainting with Masked Large Reconstruction Model (NeurIPS 2025 · similarity 0.86)
- DeCoTR: Enhancing Depth Completion with 2D and 3D Attentions (CVPR 2024 · similarity 0.86)
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.
