README.md
Browse filesThis dataset is composed of object-centric, synthetic scenes created with Blender.
As binary NumPy files, we include training images, camera-to-world poses, coarse segmentation obtained directly from SAM2, and ground-truth segmentation generated using source geometry. We also include camera intrinsics within JSON files. The camera coordinate system is that of Blender - y is up, x is right, and the camera looks down the negative z-axis.
README.md
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-nc-4.0
|
| 3 |
+
task_categories:
|
| 4 |
+
- image-to-3d
|
| 5 |
+
language:
|
| 6 |
+
- en
|
| 7 |
+
tags:
|
| 8 |
+
- gaussian-splatting
|
| 9 |
+
- 3d-reconstruction
|
| 10 |
+
- nerf
|
| 11 |
+
- photogrammetry
|
| 12 |
+
- 3d-segmentation
|
| 13 |
+
- image-segmentation
|
| 14 |
+
- splats
|
| 15 |
+
pretty_name: >-
|
| 16 |
+
Vanishing Volumes: Calibrated Scenes With Segmentation for the 3D
|
| 17 |
+
Reconstruction of Fine-Structured Objects
|
| 18 |
+
---
|