Downloads · 30 days
0
p-core
victor678/p-core
p-core is a machine learning model from victor678. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as mit.
Checkpoints and data for P-CORE: Self-Supervised Surface Consistency for Point-Based Neural Editing (ECCV 2026). Code: https://github.com/zvict/p-core
Access
Public
Updated Sep 7, 2026
Repo size
4.6 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.png2.4 GB · 46%
At a glance
- License
- mit
- Access
- Public
- Created
- Sep 3, 2026
- Updated
- Sep 7, 2026
- SHA
- 7ff96463
Model card
From the Hugging Face model README
P-CORE release assets
Checkpoints and data for P-CORE: Self-Supervised Surface Consistency for Point-Based Neural Editing (ECCV 2026). Code: https://github.com/zvict/p-core
Download everything into a checkout so the paths match what the configs expect:
hf download victor678/p-core --local-dir .
Contents
| Path | What it is |
|---|---|
checkpoints/{neural_editor,objaverse}/<scene>/model.pth | Released weights, one per benchmark scene |
checkpoints/{neural_editor/ship,objaverse/lego}/base_model.pth | Reconstruction parents those two scenes initialize from |
data/Objaverse/<scene>/start | Our processed PAPR-in-Motion Objaverse subset (canonical views) |
datasets/{neural_editor,objaverse}/<scene> | Deformed views, used only at evaluation time |
edits/<benchmark>/<scene>/points.ply | Deformed point clouds |
2dgs/ | 2DGS initial point clouds and depth supervision |
The NeRF Synthetic scenes are not redistributed here; get them from the link in
the code repository's README and place them under data/nerf_synthetic/.
- License
- mit
- Created
- Sep 3, 2026
- Updated
- Sep 7, 2026