Fix Space metadata and publish core test
Browse files
README.md
CHANGED
|
@@ -1,13 +1,20 @@
|
|
| 1 |
---
|
| 2 |
title: SimamAnything2GS
|
| 3 |
-
emoji:
|
| 4 |
-
colorFrom:
|
| 5 |
-
colorTo:
|
| 6 |
sdk: gradio
|
| 7 |
-
sdk_version: 6.27.0
|
| 8 |
-
python_version: '3.13'
|
| 9 |
app_file: app.py
|
| 10 |
-
pinned: false
|
| 11 |
---
|
| 12 |
|
| 13 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
title: SimamAnything2GS
|
| 3 |
+
emoji: ✨
|
| 4 |
+
colorFrom: blue
|
| 5 |
+
colorTo: blue
|
| 6 |
sdk: gradio
|
|
|
|
|
|
|
| 7 |
app_file: app.py
|
|
|
|
| 8 |
---
|
| 9 |
|
| 10 |
+
# SimamAnything2GS
|
| 11 |
+
|
| 12 |
+
An ingest-first workbench for turning arbitrary images and videos into a
|
| 13 |
+
reproducible Gaussian-reconstruction job. The initial release deliberately
|
| 14 |
+
exports a CPU-only 2.5D billboard preview so the upload, hashing, and artifact
|
| 15 |
+
contract can be tested without model weights or GPU allocation.
|
| 16 |
+
|
| 17 |
+
The preview is not a 3D reconstruction. Future adapters can add video frame
|
| 18 |
+
sampling, depth/pose estimation, multi-view fusion, and Gaussian optimization
|
| 19 |
+
behind the same manifest. Each adapter must report its model, runtime, input
|
| 20 |
+
hashes, and held-out-view metrics.
|