Research map / Point clouds / Shape and surface completion
Learning Efficient Point Cloud Generation for Dense 3D Object Reconstruction
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use 2D-convolutional dense point-cloud generation with joint 2D projection optimization instead of volumetric 3D ConvNet prediction | Single-image chair reconstruction · ShapeNet chair test split | 3D error, pred. → GT: 1.827 -> 1.768 | Helped |
| Use 2D-convolutional dense point-cloud generation with joint 2D projection optimization instead of volumetric 3D ConvNet prediction | Single-image chair reconstruction · ShapeNet chair test split | 3D error, GT → pred.: 2.660 -> 1.763 | Helped |
| Use 2D-convolutional dense point-cloud generation with joint 2D projection optimization instead of PTN projection-only prediction | Single-image chair reconstruction · ShapeNet chair test split | 3D error, pred. → GT: 2.181 -> 1.768 | Helped |
| Use 2D-convolutional dense point-cloud generation with joint 2D projection optimization instead of PTN projection-only prediction | Single-image chair reconstruction · ShapeNet chair test split | 3D error, GT → pred.: 2.170 -> 1.763 | Helped |
| Explicitly generate and geometrically combine point-cloud structure rather than implicitly encode 3D geometry | Single-image chair reconstruction · ShapeNet chair test split | 3D error, pred. → GT: 2.381 -> 1.768 | Helped |
| Explicitly generate and geometrically combine point-cloud structure rather than implicitly encode 3D geometry | Single-image chair reconstruction · ShapeNet chair test split | 3D error, GT → pred.: 3.019 -> 1.763 | Helped |
| Use dense point-cloud generation instead of Fan et al.'s unordered 1024-point prediction | Single-image reconstruction across 13 categories · ShapeNet 13-category test mean, single view | 3D error, pred. → GT: 1.982 -> 1.846 | Helped |
| Use dense point-cloud generation instead of Fan et al.'s unordered 1024-point prediction | Single-image reconstruction across 13 categories · ShapeNet 13-category test mean, single view | 3D error, GT → pred.: 2.146 -> 1.701 | Helped |
Closest papers by meaning
- A Point Set Generation Network for 3D Object Reconstruction From a Single Image (CVPR 2017 · similarity 0.92)
- Learning Progressive Point Embeddings for 3D Point Cloud Generation (CVPR 2021 · similarity 0.90)
- MVPNet: Multi-View Point Regression Networks for 3D Object Reconstruction from A Single Image (AAAI 2019 · similarity 0.89)
- From Image Collections to Point Clouds With Self-Supervised Shape and Pose Networks (CVPR 2020 · similarity 0.89)
- Synthesizing 3D Shapes via Modeling Multi-View Depth Maps and Silhouettes With Deep Generative Networks (CVPR 2017 · similarity 0.89)
- SPAR3D: Stable Point-Aware Reconstruction of 3D Objects from Single Images (CVPR 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.