Improve metadata and add technical documentation

README.md

@@ -3,8 +3,14 @@ license: other
 license_name: stabilityai-ai-community
 license_link: LICENSE.md
 pipeline_tag: image-to-3d
+datasets:
+- objaverse
 tags:
 - image-to-3d
+- 3d
+- reconstruction
+- relighting
+- multi-view
 - checkpoint
 inference: false
 ---
@@ -34,6 +40,27 @@ Please note: For individuals or organizations generating annual revenue of US $1
 * **Project page**: https://reli3d.jdihlmann.com/
 * **arXiv paper**: https://arxiv.org/abs/2603.19753
 
+### Technical Details
+
+#### Input Format
+Each object directory must contain a camera specification and RGBA frames:
+```text
+input_root/
+  object_name/
+    transforms.json
+    rgba/
+      0000.png
+      0010.png
+      ...
+```
+Expected `transforms.json` frame keys include `file_path`, `transform_matrix` (or `camera_transform`), and `camera_fov`.
+
+#### Outputs
+The model produces the following per object:
+- `mesh.glb`: UV-unwrapped textured mesh.
+- `illumination.hdr`: Estimated environment illumination (if predicted).
+- `run_info.json`: Metadata about the inference run.
+
 ### Files
 
 * **config.yaml**: ReLi3D inference config
@@ -64,6 +91,8 @@ python demos/reli3d/infer_from_transforms.py \
   --input-root demo_files/objects \
   --objects Camera_01 \
   --output-root outputs \
+  --num-views 4 \
+  --texture-size 256 \
   --overwrite
 ```
 
@@ -107,4 +136,4 @@ Please report any issues with the model or contact us:
 * Security issues: security@stability.ai
 * Privacy issues: privacy@stability.ai
 * License and general: https://stability.ai/license
-* Enterprise license: https://stability.ai/enterprise
+* Enterprise license: https://stability.ai/enterprise