Research map / 3D vision / Face and shape
UCLID-Net: Single View Reconstruction in Object Space
Benoit Guillard, Edoardo Remelli, Pascal Fua
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 |
|---|---|---|---|
| UCLID-Net vs DISN | single-view reconstruction · ShapeNet Core | CD-L2 (x10^3): 9.7 -> 6.3 | Improved |
| UCLID-Net vs Pixel2Mesh+ | single-view reconstruction · ShapeNet Core | CD-L2 (x10^3): 7.0 -> 6.3 | Improved |
| UCLID-Net vs DISN | single-view reconstruction · ShapeNet Core | EMD (x10^2): 2.6 -> 2.5 | Improved |
| UCLID-Net vs Pixel2Mesh+ | single-view reconstruction · ShapeNet Core | shell-IoU (%): 30 -> 37 | Improved |
| UCLID-Net vs Pixel2Mesh+ | single-view reconstruction · ShapeNet Core | F-Score (%): 95.0 -> 96.2 | Improved |
| UCLID-Net vs DISN on speaker category | single-view reconstruction · ShapeNet speaker | CD-L2 (x10^3): 10.3 -> 13.8 | Hurt |
| UCLID-Net vs DISN on rifle category | single-view reconstruction · ShapeNet rifle | shell-IoU (%): 60 -> 51 | Hurt |
| use ground-truth depth and camera pose instead of inferred (ablation) | UCLID-Net · ShapeNet car | CD-L2 and EMD | Improved |
| train on all 13 categories instead of cars only (ablation) | UCLID-Net · ShapeNet car | CD-L2 and EMD | Improved |
| use ground-truth camera pose only with inferred depth (ablation) | UCLID-Net · ShapeNet car | CD-L2 and EMD | Improved |
| add multi-view feature fusion by point-wise addition | UCLID-Net · ShapeNet chair | reconstruction quality | Improved |
| use instance normalization instead of batch normalization in ResNet18 encoder (ablation) | UCLID-Net · training | stability and performance | Improved |
Closest papers by meaning
- Learning Shape Priors for Single-View 3D Completion and Reconstruction (ECCV 2018 · similarity 0.88)
- View-Aware Geometry-Structure Joint Learning for Single-View 3D Shape Reconstruction (TPAMI 2021 · similarity 0.88)
- Fostering Generalization in Single-View 3D Reconstruction by Learning a Hierarchy of Local and Global Shape Priors (CVPR 2021 · similarity 0.88)
- Few-Shot Single-View 3-D Object Reconstruction with Compositional Priors (ECCV 2020 · similarity 0.88)
- Know Your Neighbors: Improving Single-View Reconstruction via Spatial Vision-Language Reasoning (CVPR 2024 · similarity 0.87)
- Less Decoder is More Encoder: Geometric Representation Learning from Novel View Synthesis (NeurIPS 2026 · 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.
