Research map / 3D vision / Face and shape
SIMstack: A Generative Shape and Instance Model for Unordered Object Stacks
Zoe Landgraf, Raluca Scona, Tristan Laidlow, Stephen James, Stefan Leutenegger, Andrew J. Davison
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 the learned conditional VAE prior over stack configurations (ablation) | SIMstack · SQ dataset | Chamfer distance (full surface): 0.175 -> 0.089 | Improved |
| add the learned VAE prior (ablation) | SIMstack · YCB dataset | Chamfer distance (full surface): 0.236 -> 0.112 | Improved |
| add the learned VAE prior (ablation) | SIMstack · SQ dataset | IoU: 0.635 -> 0.829 | Improved |
| add the learned VAE prior (ablation) | SIMstack · YCB dataset | IoU: 0.542 -> 0.784 | Improved |
| add the learned VAE prior (instance stability) (ablation) | SIMstack · SQ dataset | object center displacement (m): 1.112 -> 0.593 | Improved |
| add the learned VAE prior (instance stability) (ablation) | SIMstack · YCB dataset | object center displacement (m): 1.756 -> 1.216 | Improved |
| use joint shape and instance prediction versus SSCNet** shape-only | SIMstack · YCB dataset | Chamfer distance (full surface): 0.242 -> 0.112 | Improved |
| set Hough voting threshold mu to 10 instead of very large or small values (ablation) | instance segmentation · synthetic stacks | segmentation quality | Improved |
| condition on multiple views instead of TSDF fusion | SIMstack · stacks | reconstruction accuracy | Improved |
| use multi-view conditioning instead of multi-view optimisation | SIMstack · stacks | runtime: 75 s -> 5.2 s | Improved |
Closest papers by meaning
- PARTS: Unsupervised Segmentation With Slots, Attention and Independence Maximization (ICCV 2021 · similarity 0.88)
- Structured Prediction of Unobserved Voxels From a Single Depth Image (CVPR 2016 · similarity 0.87)
- 3D-RCNN: Instance-Level 3D Object Reconstruction via Render-and-Compare (CVPR 2018 · similarity 0.87)
- LASA: Instance Reconstruction from Real Scans using A Large-scale Aligned Shape Annotation Dataset (CVPR 2024 · similarity 0.86)
- GrabS: Generative Embodied Agent for 3D Object Segmentation without Scene Supervision (ICLR 2025 · similarity 0.86)
- Learning Partonomic 3D Reconstruction from Image Collections (CVPR 2025 · 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.
