adding bbox example. updating messaging (#24)

Co-authored-by: Jean-Philippe Fauconnier <jfauconnier@roblox.com>

Author: akashkgarg

.gitattributes

@@ -1 +1,3 @@
 examples/bulldozer.obj filter=lfs diff=lfs merge=lfs -text
+examples/truck-bbox.gif filter=lfs diff=lfs merge=lfs -text
+examples/scenegen.gif filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,21 +1,27 @@
 # Cube
 
 <p align="center">
-  <img src="./examples/green vintage couch.gif" width="200" height="200" style="margin: 5px;">
-  <img src="./examples/purple sword.gif" width="200" height="200" style="margin: 5px;">
-  <img src="./examples/unicorn.gif" width="200" height="200" style="margin: 5px;">
-  <img src="./examples/whale.gif" width="200" height="200" style="margin: 5px;">
-  <img src="./examples/red buggy.gif" width="200" height="200" style="margin: 5px;">
-  <img src="./examples/bulldozer.gif" width="200" height="200" style="margin: 5px;">
-  <img src="./examples/dragon.gif" width="200" height="200" style="margin: 5px;">
-  <img src="./examples/boat.gif" width="200" height="200" style="margin: 5px;">
+  <img src="./examples/green vintage couch.gif" width="150" height="150" style="margin: 5px;">
+  <img src="./examples/purple sword.gif" width="150" height="150" style="margin: 5px;">
+  <img src="./examples/unicorn.gif" width="150" height="150" style="margin: 5px;">
+  <img src="./examples/whale.gif" width="150" height="150" style="margin: 5px;">
+  <br />
+  <img src="./examples/red buggy.gif" width="150" height="150" style="margin: 5px;">
+  <img src="./examples/bulldozer.gif" width="150" height="150" style="margin: 5px;">
+  <img src="./examples/dragon.gif" width="150" height="150" style="margin: 5px;">
+  <img src="./examples/boat.gif" width="150" height="150" style="margin: 5px;">
 </p>
 
-<div align="center">
-  <a href=https://corp.roblox.com/newsroom/2025/03/introducing-roblox-cube target="_blank"><img src=https://img.shields.io/badge/Blog-000000.svg height=22px></a>
-  <a href=https://huggingface.co/Roblox/cube3d-v0.1 target="_blank"><img src=https://img.shields.io/badge/%F0%9F%A4%97%20Models-d96902.svg height=22px></a>
-  <a href= target="_blank"><img src=https://img.shields.io/badge/Report-b5212f.svg?logo=arxiv height=22px></a>
-</div>
+<p align="center">
+  <a href=https://corp.roblox.com/newsroom/2025/03/introducing-roblox-cube target="_blank">Blog</a>&nbsp; | 
+  <a href=https://huggingface.co/Roblox/cube3d-v0.1 target="_blank">Models on Hugging Face</a>&nbsp; | 
+  <a href= target="_blank">Technical Report</a>
+</p>
+<br>
+
+We are unlocking the power of generating 3D assets and enhancing creativity for all artists. Our latest version of Cube 3D is now accessible to individuals, creators, researchers and businesses of all sizes so that they can experiment, innovate and scale their ideas responsibly.
+
+This release includes model weights and starting code for open source pre-trained text-to-shape model.
 
 ## Get Started with Cube3D
 
@@ -31,7 +37,7 @@ source cube3d/bin/activate
 pip install -e .[meshlab]
 ```
 
-> **Cuda**: If you are using a Windows machine, you may need to install the [CUDA](https://developer.nvidia.com/cuda-downloads) toolkit as well as the torch with cuda support via `pip install torch --index-url https://download.pytorch.org/whl/cu124 --force-reinstall`
+> **CUDA**: If you are using a Windows machine, you may need to install the [CUDA](https://developer.nvidia.com/cuda-downloads) toolkit as well as the torch with cuda support via `pip install torch --index-url https://download.pytorch.org/whl/cu124 --force-reinstall`
 
 > **Note**:  The `[meshlab]` is an optional dependency and can be removed by simply running `pip install -e .` for better compatibility but mesh simplification will be disabled.
 
@@ -52,7 +58,11 @@ huggingface-cli download Roblox/cube3d-v0.1 --local-dir ./model_weights
 To generate 3D models using the downloaded models simply run:
 
 ```bash
-python3 -m cube3d.generate --gpt-ckpt-path model_weights/shape_gpt.safetensors --shape-ckpt-path model_weights/shape_tokenizer.safetensors --prompt "sleek vintage green couch with clean lines and velvet material" --fast-inference
+python3 -m cube3d.generate \
+            --gpt-ckpt-path model_weights/shape_gpt.safetensors \
+            --shape-ckpt-path model_weights/shape_tokenizer.safetensors \
+            --fast-inference \
+            --prompt "sleek vintage green couch with clean lines and velvet material"
 ```
 
 The output will be an `.obj` file saved in the specified `output` directory.
@@ -67,14 +77,16 @@ and save it as `turntable.gif` in the specified `output` directory.
 To tokenize a 3D shape into token indices and reconstruct it back, you can use the following command:
 
 ```bash
-python3 -m cube3d.vq_vae_encode_decode --shape-ckpt-path model_weights/shape_tokenizer.safetensors --mesh-path ./outputs/output.obj
+python3 -m cube3d.vq_vae_encode_decode \
+            --shape-ckpt-path model_weights/shape_tokenizer.safetensors \
+            --mesh-path ./input_mesh.obj
 ```
 
 This will process the `.obj` file located at `./outputs/output.obj` and prints the tokenized representation as well as exports the mesh reconstructed from the token indices.
 
 ### Code Usage
 
-We designed a minimalist API that allows you to use this repo as a Python library,
+We designed a minimalist API that allows you to use this repo as a Python library:
 
 ```python
 import torch
@@ -100,6 +112,18 @@ vertices, faces = mesh_v_f[0][0], mesh_v_f[0][1]
 trimesh.Trimesh(vertices=vertices, faces=faces).export("output.obj")
 ```
 
+## Coming Soon
+
+Bounding box shape conditioning: 
+<p align="center">
+  <img src="./examples/truck-bbox.gif" width="200" height="200" style="margin: 5px;">
+</p>
+
+Scene Generation: 
+<p align="center">
+  <img src="./examples/scenegen.gif" style="margin: 5px;">
+</p>
+
 ## Acknowledgements
 
 We would like to thank the contributors to [TRELLIS](https://github.com/microsoft/TRELLIS), [CraftsMan3D](https://github.com/wyysf-98/CraftsMan3D), [threestudio](https://github.com/threestudio-project/threestudio), [Hunyuan3D-2](https://github.com/Tencent/Hunyuan3D-2), [minGPT](https://github.com/karpathy/minGPT), [dinov2](https://github.com/facebookresearch/dinov2), [OptVQ](https://github.com/zbr17/OptVQ), [1d-tokenizer](https://github.com/bytedance/1d-tokenizer)