py-soma-x 0.1.3__tar.gz

py_soma_x-0.1.3/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

py_soma_x-0.1.3/MANIFEST.in

@@ -0,0 +1,6 @@
+include LICENSE
+include README.md
+recursive-include soma *.py
+prune assets
+prune tools
+prune tests

py_soma_x-0.1.3/PKG-INFO

@@ -0,0 +1,403 @@
+Metadata-Version: 2.4
+Name: py-soma-x
+Version: 0.1.3
+Summary: SOMA body layer
+Home-page: https://github.com/NVlabs/SOMA-X
+Author: NVIDIA Corporation
+License: Apache-2.0
+Project-URL: Bug Tracker, https://github.com/NVlabs/SOMA-X/issues
+Project-URL: Source Code, https://github.com/NVlabs/SOMA-X
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: Science/Research
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Programming Language :: Python :: 3
+Classifier: Topic :: Scientific/Engineering
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: numpy
+Requires-Dist: scipy
+Requires-Dist: torch
+Requires-Dist: trimesh
+Requires-Dist: warp-lang
+Requires-Dist: rtree
+Requires-Dist: cholespy
+Requires-Dist: huggingface_hub
+Provides-Extra: smpl
+Requires-Dist: smplx; extra == "smpl"
+Provides-Extra: anny
+Requires-Dist: anny; extra == "anny"
+Requires-Dist: Pillow; extra == "anny"
+Provides-Extra: dev
+Requires-Dist: pytest; extra == "dev"
+Requires-Dist: ruff; extra == "dev"
+Requires-Dist: pre-commit; extra == "dev"
+Provides-Extra: demo
+Requires-Dist: pyrender; extra == "demo"
+Requires-Dist: tqdm; extra == "demo"
+Requires-Dist: imageio[ffmpeg]; extra == "demo"
+Provides-Extra: all
+Requires-Dist: smplx; extra == "all"
+Requires-Dist: anny; extra == "all"
+Requires-Dist: Pillow; extra == "all"
+Dynamic: license-file
+
+
+
+<p align="center">
+  <img src="./assets/images/banner.png" alt="Banner" width="100%">
+</p>
+
+
+## Overview
+
+Parametric human body models, including SMPL, SMPL-X, MHR, Anny, and GarmentMeasurements, are central to a wide range of tasks in human reconstruction, animation, and simulation. However, these models are inherently incompatible: each defines its own mesh topology, joint hierarchy, and parameterization, precluding seamless integration. As a result, leveraging complementary strengths across models (such as combining Anny’s age-range control with SMPL-based motion data) necessitates bespoke adapters for every model pair, hindering interoperability and limiting practical applications.
+
+We present **SOMA**—a canonical body topology and rig that acts as a universal pivot for all supported parametric human body models. Instead of replacing existing models, **SOMA unifies them** by mapping their diverse rest shapes onto a single, shared representation. This approach allows any supported identity model to be animated with a unified animation pipeline, eliminating the need for custom adapters or model-specific retargeting. With SOMA, you can mix and match identity sources and pose data at inference time without additional engineering. The entire pipeline remains end-to-end differentiable and GPU-accelerated via NVIDIA Warp.
+
+See SOMA in action: 
+
+<p align="center">
+  <img src="assets/images/soma-in-action.gif" alt="SOMA in Action" width="1000"/>
+</p>
+
+## Supported Identity Models
+
+SOMA currently supports five distinct identity models, each offering unique capabilities:
+
+1. [MHR](https://github.com/facebookresearch/MHR): The default identity model in SOMA, providing high-fidelity body shape representation.
+2. [Anny](https://github.com/naver/anny): Particularly well-suited for modeling children, broadening applicability to younger subjects.
+3. [SMPL-Family](https://smpl.is.tue.mpg.de/): Supports both SMPL and SMPL-X models, enabling interoperability with established standards in the field.
+4. **SOMA-shape**: A proprietary PCA-based model developed as part of this project, designed to offer SMPL-like functionality with 128 PCA coefficients for identity representation. 
+5. [GarmentMeasurement](https://github.com/mbotsch/GarmentMeasurements): A PCA-based identity model trained on the CASERS dataset, suitable for specialized use cases involving garment fitting and measurement.
+
+We welcome community contributions to extend support for additional identity models.
+
+## Unified Pose Correctives (Beta)
+Thanks to SOMA's unified framework, pose-dependent corrective deformations that mitigate LBS artifacts are seamlessly available for all supported identity models, including those that do not provide correctives themselves (e.g., Anny and GarmentMeasurement).
+<p align="center">
+  <img src="assets/images/soma_correctives.gif" alt="SOMA Pose Correctives" width="800"/>
+</p>
+
+
+
+## Installation
+
+### Install from PyPI
+
+```bash
+pip install py-soma-x
+```
+
+With optional extras:
+```bash
+pip install "py-soma-x[smpl]"   # SMPL/SMPL-X support
+pip install "py-soma-x[anny]"   # Anny support
+```
+
+Assets are automatically downloaded from HuggingFace on first use (cached in `~/.cache/huggingface/hub/`).
+
+> **Note:** SMPL/SMPL-X requires `chumpy`, which must be installed separately:
+> ```bash
+> pip install --no-build-isolation chumpy
+> ```
+> If that fails, install from source:
+> ```bash
+> pip install --no-build-isolation git+https://github.com/mattloper/chumpy@580566eafc9ac68b2614b64d6f7aaa8
+> ```
+>
+> SMPL/SMPL-X model files (`SMPL_NEUTRAL.pkl`, `SMPLX_NEUTRAL.npz`) require a separate license and must be downloaded from [SMPL](https://smpl.is.tue.mpg.de/) / [SMPL-X](https://smpl-x.is.tue.mpg.de/). Pass the model path explicitly:
+> ```python
+> soma = SOMALayer(
+>     identity_model_type="smpl",
+>     identity_model_kwargs={"model_path": "/path/to/SMPL_NEUTRAL.pkl"},
+> )
+> ```
+
+<details>
+
+<summary>Developer installation (clone with Git LFS)</summary>
+
+### Clone with Git LFS (Required for Assets)
+This repository uses Git LFS for large asset files (e.g., assets/Nova_neutral.npz). You must install Git LFS to download the actual data; otherwise, you will encounter file loading errors.
+
+1. Install Git LFS (if not installed):
+````bash
+git lfs install
+````
+
+2. Clone and Pull Data:
+```bash
+git clone https://github.com/NVlabs/SOMA-X.git
+cd SOMA-X
+git lfs pull
+```
+_(If you already cloned the repo, just run `git lfs pull` to fetch the missing assets.)_
+
+### Prepare Python environment
+
+**Linux:**
+```bash
+pip install uv
+uv venv .venv
+source .venv/bin/activate   # or: . .venv/bin/activate
+# Install PyTorch with CUDA — adjust the version (cu124, cu126, cu130, …)
+# to match your GPU and driver. See https://pytorch.org/get-started/locally/
+uv pip install torch --index-url https://download.pytorch.org/whl/cu124
+uv pip install ".[dev]"
+```
+
+**Windows (PowerShell):**
+```powershell
+pip install uv
+uv venv .venv
+Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser   # one-time setup
+.\.venv\Scripts\activate
+# Install PyTorch with CUDA — adjust the version (cu124, cu126, cu130, …)
+# to match your GPU and driver. See https://pytorch.org/get-started/locally/
+uv pip install torch --index-url https://download.pytorch.org/whl/cu124
+uv pip install ".[dev]"
+```
+Then run tests: `pytest tests/ -v`.  
+
+### Optional Dependencies
+
+**For SMPL and SMPLX support:**
+
+```bash
+uv pip install ".[smpl]"
+pip install --no-build-isolation chumpy
+```
+_NOTE: `chumpy` (required by `smplx` at runtime) has a broken PyPI build and must be installed with `--no-build-isolation`. If that fails, install from source: `pip install --no-build-isolation git+https://github.com/mattloper/chumpy@580566eafc9ac68b2614b64d6f7aaa8`_
+
+You also need to download `SMPL_NEUTRAL.pkl` or `SMPLX_NEUTRAL.npz` separately:
+1. Visit the [SMPL](https://smpl.is.tue.mpg.de/) or [SMPLX](https://smpl-x.is.tue.mpg.de/) website. 
+2. Register and download the SMPL (v1.1.0 for Python) or [SMPL-X](https://download.is.tue.mpg.de/download.php?domain=smplx&sfile=smplx_lockedhead_20230207.zip) (with removed head bun) model files. 
+3. Extract and copy `SMPL_NEUTRAL.pkl` to `./assets/SMPL/SMPL_NEUTRAL.pkl` and `SMPLX_NEUTRAL.npz` to `./assets/SMPLX/SMPLX_NEUTRAL.npz`. 
+
+**Note:** The SMPL models are subject to a separate license and cannot be redistributed with this repository.
+
+**For [Anny](https://github.com/naver/anny) support:**
+```bash
+uv pip install ".[anny]"
+```
+
+**For [GarmentMeasurement](https://github.com/mbotsch/GarmentMeasurements) support:**
+```bash
+git clone https://github.com/mbotsch/GarmentMeasurements 
+python tools/convert_gm_pca_to_npz.py ./GarmentMeasurements/data/pca/point.pca assets/GarmentMeasurements/point.npz
+rm -rf GarmentMeasurements 
+```
+</details>
+
+
+## Usage
+
+```python
+import torch
+from soma import SOMALayer
+
+# Initialize the layer — assets are auto-downloaded from HuggingFace
+soma = SOMALayer(
+    identity_model_type="mhr",  # or "soma" "smpl", "smplx", "anny", "garment"
+    device="cuda"
+)
+
+# Or use a local assets directory
+# soma = SOMALayer(data_root="./assets", identity_model_type="mhr", device="cuda")
+
+# Forward pass
+# poses: (B, num_joints, 3)
+# identity: (B, num_coeffs)
+# scale_params: (B, num_scales) - Optional, depending on model type (required for MHR)
+output = soma(poses, identity, scale_params=scale_params)
+vertices = output["vertices"]
+```
+
+## Running the Demo
+
+Install the demo environment (includes pyrender, tqdm, imageio with ffmpeg for video output):
+
+```bash
+uv pip install ".[demo]"
+```
+
+If you want to run all identity models (soma, mhr, anny, smpl, smplx, garment), install the full set and use the same build steps as for tests:
+
+```bash
+uv pip install -e ".[all,demo]"
+pip install --no-build-isolation chumpy
+```
+
+Then run the demo script:
+
+```bash
+# Run all models (default: soma, mhr, anny, smpl, smplx, garment)
+python tools/demo_soma_vis.py --data-root ./assets --output-dir ./out
+
+# Run specific models only
+python tools/demo_soma_vis.py --data-root ./assets --output-dir ./out --identity-model-type soma, mhr, smplx
+
+# Run a single model
+python tools/demo_soma_vis.py --data-root ./assets --output-dir ./out --identity-model-type anny
+
+# Run MHR with random shapes
+python tools/demo_soma_vis.py --data-root ./assets --output-dir ./out --identity-model-type mhr --random-shape
+```
+
+This will generate example animation videos for the selected models in the `out/` directory.
+
+**Demo Options:**
+- `--identity-model-type`: Comma-separated list of models to use (options: `soma`, `mhr`, `anny`, `smplx`, `smpl`, `garment`, default: `soma,mhr,anny,smpl,smplx,garment`)
+- `--random-shape`: Generate random body shapes instead of using neutral shapes
+- `--motion-file`: Path to custom motion file (default: `assets/ROM5.npy`)
+- `--image-size`: Render resolution (default: 1920)
+- `--device`: Device to use (default: `cuda:0`)
+
+## Conversion of pose parameters from other models to SOMA
+We provide conversion tools for converting from SMPL and MHR pose parameters to SOMA.
+Both tools use `PoseInversion.fit()`, which supports two complementary solvers — both initialized by a single-pass skeleton transfer fit for fast convergence:
+
+- **Analytical** (default): iterative inverse-LBS with Newton-Schulz refinement. Extremely fast (~1200 FPS) with comparable accuracy.
+- **Autograd FK**: 6D rotation optimization by backpropagating FK + LBS. Slow but controllable (e.g. extra weights on extremities).
+
+The two can be combined: the analytical solve warm-starts autograd refinement — best of both worlds.
+
+### SMPL to SOMA
+
+<img src="assets/images/smpl2soma.gif" alt="SMPL to SOMA conversion" width="400"/>
+
+```bash
+# Convert SMPL animation to SOMA (renders comparison video)
+python -m tools.smpl2soma
+
+# Export SOMA poses as .npz
+python -m tools.smpl2soma --output-npz out/smpl_soma.npz
+
+# Tune analytical iterations (defaults: --body-iters 2 --full-iters 1)
+python -m tools.smpl2soma --body-iters 3 --full-iters 1 --batch-size 64
+
+# Analytical + autograd FK refinement (best accuracy)
+python -m tools.smpl2soma --body-iters 2 --full-iters 1 --autograd-iters 10
+```
+
+**Benchmark** (402 SMPL frames, RTX 5000 Ada):
+
+| Method | Speed | Mean | Median | Max |
+|---|---|---|---|---|
+| Analytical (body=2, full=1) — **default** | **1279 FPS** | 0.65 cm | 0.52 cm | 17.8 cm |
+| Autograd FK (10 iters, lr=5e-3) | 199 FPS | 1.04 cm | 0.97 cm | 18.1 cm |
+| Autograd FK (100 iters) | 18 FPS | 0.49 cm | 0.39 cm | 16.8 cm |
+
+### MHR to SOMA
+
+<img src="assets/images/mhr2soma.gif" alt="MHR to SOMA conversion" width="400"/>
+
+For [SAM 3D Body](https://huggingface.co/datasets/facebook/sam-3d-body-dataset) or similar MHR-format data.
+
+```bash
+# Convert a directory of SAM 3D Body parquet files
+python -m tools.mhr2soma --input path/to/sam_3d_body/data/coco_train
+
+# Convert and export as .npz
+python -m tools.mhr2soma --input path/to/parquet_dir --output-npz out/mhr_soma.npz
+
+# Tune analytical iterations (defaults: --body-iters 2 --full-iters 1)
+python -m tools.mhr2soma --input path/to/parquet_dir --max-samples 100 --body-iters 3
+
+# Analytical + autograd FK refinement (best accuracy)
+python -m tools.mhr2soma --input path/to/parquet_dir --autograd-iters 10
+```
+
+**Benchmark** (200 SAM 3D Body samples, RTX 5000 Ada):
+
+| Method | Speed | Mean | Median | Max |
+|---|---|---|---|---|
+| Analytical (body=2, full=1) — **default** | **342 FPS** | 0.61 cm | 0.34 cm | 14.8 cm |
+| Autograd FK (10 iters, lr=5e-3) | 161 FPS | 1.05 cm | 0.76 cm | 13.5 cm |
+| Autograd FK (100 iters) | 16 FPS | 0.48 cm | 0.22 cm | 13.3 cm |
+
+> **Note:** The `mhr2soma` tool's end-to-end throughput (~50 samp/s) is dominated by MHR identity model evaluation, not SOMA inversion. The MHR TorchScript model is called twice per sample (once to produce the rest shape, once for posed vertices). The SOMA inversion itself runs at 342 FPS.
+
+### AMASS dataset to SOMA
+
+Convert [AMASS](https://amass.is.tue.mpg.de/) motion sequences (SMPL format `.npz` files) to SOMA.
+
+> **Prerequisites:** Download the AMASS dataset from [amass.is.tue.mpg.de](https://amass.is.tue.mpg.de/) and place `SMPL_NEUTRAL.pkl` in `assets/SMPL/` (see SMPL installation above).
+
+```bash
+# Single file — converts and renders a comparison video
+python -m tools.convert_amass_to_soma --input path/to/amass_sequence.npz
+
+# Single file — export .npz only (skip rendering)
+python -m tools.convert_amass_to_soma --input path/to/amass_sequence.npz --output-npz out/soma.npz --no-render
+
+# Batch convert entire dataset (mirrors folder structure)
+python -m tools.convert_amass_to_soma --input-dir /data/amass/ --output-dir out/amass_soma/
+
+# Shuffle file order (useful when running multiple workers in parallel)
+python -m tools.convert_amass_to_soma --input-dir /data/amass/ --output-dir out/amass_soma/ --shuffle
+
+# Tune analytical iterations
+python -m tools.convert_amass_to_soma --input path/to/seq.npz --body-iters 3 --full-iters 1
+
+# Analytical + autograd FK refinement (best accuracy)
+python -m tools.convert_amass_to_soma --input path/to/seq.npz --autograd-iters 10
+```
+
+The output `.npz` files contain:
+- `poses`: `(N, J, 3)` rotation vectors per joint
+- `root_translation`: `(N, 3)` root position in meters
+- `joint_names`: list of SOMA joint names
+- `per_vertex_error`: `(N, V)` reconstruction error per vertex
+- `identity_coeffs` / `scale_params`: identity parameters used
+
+
+**Benchmark** (A100):
+
+| Method | Speed | Mean | Median | Max |
+|---|---|---|---|---|
+| Analytical (body=2, full=1) — **default** | **17393 FPS** | 0.53 cm | 0.32 cm | 8.8 cm |
+| Autograd FK (10 iters, lr=5e-3) | 435 FPS | 0.78 cm | 0.64 cm | 8.8 cm |
+
+
+## Related Human(oid) Work at NVIDIA
+SOMA is part of a larger effort to enable human animation, robotics, physical AI, and other applications.  
+
+Check out these related works that already support SOMA:  
+* [GEM](https://github.com/NVlabs/GEM-X) - SOMA-based video pose estimation. 
+* [Kimodo](https://github.com/nv-tlabs/kimodo) - SOMA-based controllable text-to-motion generation method for **human(oid)s**.  
+* [BONES-SEED Dataset](https://huggingface.co/datasets/bones-studio/seed) - a large scale human(oid) motion capture dataset in SOMA format. Also provides retargeted G1 data.  
+* [SOMA Retargeter](https://github.com/NVIDIA/soma-retargeter) - for SOMA to G1 retargeting. 
+* [ProtoMotion](https://github.com/NVlabs/ProtoMotions) - simulation and learning framework for training physically simulated digital human(oid)s
+* [GEAR SONIC](https://github.com/NVlabs/GR00T-WholeBodyControl) - a humanoid behavior foundation model. (coming soon)
+
+
+
+## Citation
+If you use this code in your work, please cite:
+
+```bibtex
+@article{soma2026,
+  title={SOMA: Unifying Parametric Human Body Models},
+  author={[AUTHORS TO BE FILLED]},
+  eprint={2603.XXXXX},
+  archivePrefix={arXiv},
+  year={2026},
+  url={https://arxiv.org/abs/2603.XXXXX},
+}
+```
+
+## Acknowledgements 
+- [SMPL-Body](https://smpl.is.tue.mpg.de/bodylicense.html) was used to create an interpolator between SMPL and SOMA mesh topologies, courtesy of the Max Planck Institute for Intelligent Systems. 
+- [MHR](https://github.com/facebookresearch/MHR) was used to learn the pose corrective model. 
+- [Anny](https://github.com/naver/anny) for [WARP](https://github.com/NVIDIA/warp)-based sparse linear blend skinning.  
+- [GarmentMeasurement](https://github.com/mbotsch/GarmentMeasurements) was used to augment the data in our shape model. 
+
+
+## License
+
+This codebase is licensed under [Apache-2.0](LICENSE). 
+
+This project will download and install additional third-party open source software projects. Review the license terms of these open source projects before use.