Datasets:

Appleartsstudios
/

Full_Length_Motion_Capture_Dataset

Effective Date: 31 July 2026
Dataset Name: Apple Arts Studios Full-Length Motion Capture Dataset
Repository: Appleartsstudios/Full_Length_Motion_Capture_Dataset
Licensor: Apple Arts Studios LLP, trading as Apple Arts Studios
Contact: hello@appleartsstudios.com
Website: https://www.appleartsstudios.com/

IMPORTANT NOTICE

This Apple Arts Studios Motion Dataset License Agreement (the “Agreement”) is a legally binding agreement between Apple Arts Studios LLP, trading as Apple Arts Studios (the “Licensor”), and the individual or legal entity that accesses or uses the Dataset (the “Licensee”).

By clicking an access or acceptance button, requesting access, downloading, copying, accessing through an API or script, or otherwise using any part of the Dataset, Licensee confirms that Licensee:

has read and understood this Agreement;
agrees to be bound by this Agreement; and
has authority to accept this Agreement personally or on behalf of the organisation represented by Licensee.

If Licensee does not agree to every term of this Agreement, Licensee must not access, download, copy, or use the Dataset.

This Agreement permits only the Non-Commercial Uses expressly described below. Any Commercial Use requires a separate written commercial licence from Licensor.

1. DEFINITIONS

1.1 “Dataset”

“Dataset” means the Apple Arts Studios Full-Length Motion Capture Dataset made available through the Repository, including, where provided:

OptiTrack Motive Proportional BVH motion files;
SOMA Uniform BVH motion files;
Unitree G1 retargeted CSV files;
Original motion sequences;
Mirrored motion sequences;
metadata, labels, descriptions, annotations, identifiers, and documentation;
full-length continuous human motion capture sequences;
folder structures, category hierarchies, taxonomies, schemas, and file-naming systems;
corrections, replacements, and updates; and
any copy, portion, extraction, conversion, or substantially equivalent representation of the foregoing.

1.2 “Raw Data”

“Raw Data” means the Dataset files and any data derived from them that preserves substantially the same motion information, including converted, retargeted, reformatted, resampled, mirrored, normalised, reconstructed, or extracted motion data.

Raw Data includes, without limitation, equivalent or substantially equivalent data represented as BVH, CSV, FBX, C3D, NPZ, NPY, JSON, JSONL, Parquet, animation curves, skeletal transforms, joint rotations, joint positions, trajectories, keyframes, or another machine-readable motion format.

1.3 “Documentation”

“Documentation” means the Dataset card, README files, metadata documentation, schemas, taxonomies, category definitions, technical notes, file descriptions, and other materials accompanying the Dataset.

1.4 “Results”

“Results” means outputs created through a permitted use of the Dataset, including:

academic publications;
research findings;
statistical analyses;
benchmark reports;
plots and visualisations;
rendered videos and demonstrations;
trained models;
model weights;
predictions;
classifications;
pose estimates;
control policies; and
non-commercial software.

Results do not include Raw Data, a Competing Dataset, or any output from which Raw Data or substantially equivalent motion sequences can reasonably be extracted, reconstructed, or regenerated.

1.5 “Non-Commercial Use”

“Non-Commercial Use” means use undertaken solely for:

academic or scientific research;
education or personal study;
non-commercial benchmarking and evaluation;
publication of non-commercial research; or
internal proof-of-concept experimentation that is not connected to a commercial product, paid service, client engagement, production deployment, or revenue-generating activity.

1.6 “Commercial Use”

“Commercial Use” means any use that is intended for or connected with commercial advantage, monetary compensation, revenue generation, paid services, product development, production deployment, client work, advertising, fundraising deliverables, licensing, resale, or business operations.

Commercial Use includes, without limitation:

use in a commercial product or service;
use for paid client work or work-for-hire;
use in production software, games, films, applications, simulations, robots, or animation pipelines;
use to train or improve a model offered commercially;
use by a commercial entity for product research or product development;
use to create assets sold or licensed to third parties; and
use intended to reduce or replace the need to purchase or license motion data.

1.7 “Competing Dataset”

“Competing Dataset” means any dataset, data library, motion library, synthetic dataset, data service, or collection substantially similar to the Dataset in scope, content, functionality, or intended use and made available to any third party.

1.8 “Authorised User”

“Authorised User” means a person who has independently accepted this Agreement and, where access controls are used, has been granted access by Licensor.

2. ELIGIBILITY AND AUTHORITY

2.1 Eligible Use

The Dataset may be used under this Agreement only for Non-Commercial Use.

2.2 Acceptance on Behalf of an Organisation

A person accepting this Agreement on behalf of a university, research institute, company, laboratory, or other organisation represents and warrants that the person has legal authority to bind that organisation.

In that case, “Licensee” means the organisation.

2.3 Commercial Entities

A commercial entity is not granted rights for product research, product development, production use, paid services, client work, commercial model training, or any other Commercial Use under this Agreement.

A commercial entity must obtain a separate written licence from Licensor before undertaking any such use.

3. LICENCE GRANT

Subject to Licensee’s continuing compliance with this Agreement, Licensor grants Licensee a limited, revocable, non-exclusive, non-transferable, non-sublicensable, worldwide licence to:

access and download the Dataset;
store reasonable internal copies for authorised Non-Commercial Use;
process, analyse, transform, convert, resample, retarget, or computationally evaluate the Dataset;
use the Dataset to train, fine-tune, test, benchmark, or evaluate machine-learning, computer-vision, animation-research, or robotics systems solely for Non-Commercial Use;
create Results; and
publish or display Results only as permitted by Sections 8 and 9.

No right is granted except as expressly stated in this Agreement.

4. PERMITTED NON-COMMERCIAL USES

Subject to all restrictions in this Agreement, permitted uses include:

academic research;
scientific research;
educational coursework and teaching;
personal study;
motion analysis;
pose-estimation research;
action-recognition research;
motion-tracking research;
behaviour-cloning research;
imitation-learning research;
embodied-AI research;
vision-language-action research;
robotics-policy research;
non-commercial benchmarking;
non-commercial model evaluation;
non-commercial internal prototyping; and
rendered demonstrations that do not expose or distribute Raw Data.

5. COMMERCIAL LICENCE REQUIRED

This Agreement does not authorise Commercial Use.

Licensee must obtain prior written permission and a separate commercial licence from Licensor before using the Dataset or Results for any Commercial Use.

Commercial licensing requests must be sent to:

Apple Arts Studios
Email: hello@appleartsstudios.com
Website: https://www.appleartsstudios.com/

Repository access, approval of an access request, silence, informal discussion, or receipt of the Dataset does not constitute commercial permission.

Commercial permission is valid only when granted in a separate written agreement by Licensor.

6. RESTRICTIONS

Licensee must not, and must not enable, assist, or permit another person or entity to do any of the following.

6.1 No Redistribution, Resale, or Transfer

Licensee must not:

sell, resell, rent, lease, lend, license, sublicense, distribute, transfer, publish, or share the Dataset or Raw Data;
provide the Dataset or Raw Data to an affiliate, client, customer, contractor, collaborator, student, or third party who is not an Authorised User;
upload or mirror the Dataset or Raw Data to Hugging Face, GitHub, cloud storage, file-sharing services, data marketplaces, or another public or third-party platform;
create or distribute a public fork, mirror, backup, archive, torrent, or duplicate repository;
include Raw Data in software packages, model repositories, containers, applications, checkpoints, downloadable products, or public demonstrations; or
make Raw Data accessible through an API, website, service, model endpoint, application, or other interface.

A collaborator must independently obtain authorised access from Licensor before receiving or using the Dataset.

6.2 No Competing Dataset

Licensee must not use the Dataset, Raw Data, or Results to:

create, improve, validate, enrich, label, augment, advertise, or distribute a Competing Dataset;
create a commercially substitutable synthetic motion dataset;
create a motion library or animation library for third-party use;
publish converted or retargeted versions of the Dataset;
combine the Dataset with other data for the purpose of releasing a new motion dataset; or
reduce or eliminate a third party’s need to license the Dataset from Licensor.

6.3 Restrictions on Generative Motion Systems

Licensee must not use the Dataset, Raw Data, or Results to train, fine-tune, condition, evaluate, or improve a model whose primary purpose or substantial use is to generate:

motion-capture data;
skeletal motion sequences;
animation files;
humanoid-motion datasets;
synthetic human-motion data; or
other outputs that function as a substitute for the Dataset.

This restriction does not prohibit permitted Non-Commercial Use of models that produce non-dataset outputs such as classifications, predictions, pose estimates, tracking outputs, control signals, or robotics policies, provided that Raw Data cannot reasonably be recovered from those outputs.

6.4 No Commercial Use Without Permission

Without a separate written commercial licence, Licensee must not:

use the Dataset or Results in a commercial product or service;
perform paid work, consulting, outsourcing, development, or model training using the Dataset;
use the Dataset in a production system;
use the Dataset in a commercial robot, game, film, simulation, application, virtual-production pipeline, or animation product;
sell, license, monetise, or commercially distribute Results; or
use the Dataset for product development, commercial research and development, or client deliverables.

6.5 No Re-identification

Licensee must not attempt to identify or re-identify any performer, actor, subject, or individual represented in the Dataset.

6.6 No Misrepresentation

Licensee must not:

claim ownership or original authorship of the Dataset;
represent that Licensee captured or produced the Dataset;
remove, hide, alter, or obscure attribution or provenance information; or
misrepresent Original and Mirrored motion sequences as independently captured performances.

6.7 Prohibited and Harmful Uses

Licensee must not use the Dataset or Results:

for an unlawful purpose;
to create deceptive or malicious deepfakes;
for malicious surveillance, tracking, harassment, or impersonation;
to discriminate unlawfully against a person or group;
to violate privacy, publicity, personality, labour, contractual, or intellectual-property rights;
to develop, train, test, or operate weapons systems; or
in violation of applicable sanctions or export-control laws.

6.8 High-Risk and Safety-Critical Use

Licensee must not use the Dataset or Results in safety-critical or hazardous deployment, including medical equipment, critical infrastructure, weapons, or hazardous autonomous systems, without a separate written agreement from Licensor and appropriate independent safety validation.

7. ACCESS CONTROL AND DATA SECURITY

Licensee must:

limit access to Authorised Users who require access for the permitted purpose;
use reasonable technical and organisational safeguards against unauthorised access, copying, disclosure, or loss;
keep internal repositories and storage locations private;
prevent credentials, API tokens, and access links from being shared;
promptly remove access when a person is no longer authorised; and
notify Licensor promptly after discovering unauthorised access, disclosure, public upload, or distribution.

A notice under this Section must be sent to hello@appleartsstudios.com.

Licensee is responsible for breaches caused by persons who obtain access through Licensee.

8. OWNERSHIP AND INTELLECTUAL PROPERTY

8.1 Licensor Ownership

Licensor retains all right, title, and interest in and to:

the Dataset;
Raw Data;
Documentation;
Original and Mirrored sequences;
metadata and annotations;
taxonomy and category structure;
file-naming conventions;
database structure and compilation;
converted, reformatted, resampled, or retargeted representations of Raw Data; and
all associated copyright, database, contractual, and proprietary rights.

The Dataset is licensed, not sold.

All rights not expressly granted are reserved by Licensor.

8.2 Licensee Ownership of Results

Subject to compliance with this Agreement, Licensee retains ownership of original Results created by Licensee.

Ownership of Results does not grant Licensee any right to:

own the Dataset or Raw Data;
redistribute or sublicense the Dataset or Raw Data;
commercially exploit the Dataset without a commercial licence;
release a Competing Dataset; or
distribute a Result from which Raw Data can reasonably be extracted or reconstructed.

8.3 No Implied Rights

No licence is granted to any trademark, trade name, logo, brand, patent, performer identity, or other right of Licensor except the limited attribution use expressly permitted in Section 9.

9. ATTRIBUTION AND CITATION

Attribution is mandatory for every public Result derived from the Dataset.

9.1 Required General Credit

Use the following credit:

Motion Data by Apple Arts Studios — https://www.appleartsstudios.com/

9.2 Academic Papers and Technical Reports

Include the credit in the acknowledgements, dataset, methods, or data-availability section.

Recommended wording:

This work uses the Apple Arts Studios Motion Dataset. Motion Data by Apple Arts Studios: https://www.appleartsstudios.com/

9.3 Models, Software, and Repositories

Include the following notice in the model card, repository README, technical documentation, or equivalent prominent location:

Training data includes motion data from the Apple Arts Studios Motion Dataset. Use of the underlying Dataset is governed by the Apple Arts Studios Motion Dataset License Agreement.

9.4 No Endorsement

Attribution does not imply that Licensor endorses Licensee, Licensee’s organisation, research, conclusions, model, software, or product.

Licensee must not use Licensor’s logo or branding except with prior written permission.

Failure to provide required attribution is a material breach of this Agreement.

10. PUBLICATIONS, MODELS, AND PUBLIC RESULTS

Licensee may publish academic papers, research reports, benchmark results, statistics, plots, visualisations, rendered demonstrations, models, and non-commercial software only when:

the publication or Result is created through a permitted Non-Commercial Use;
Raw Data is not attached, embedded, exposed, or made downloadable;
the publication does not contain sufficient joint values, transforms, coordinates, rotations, trajectories, frames, or other information to reconstruct a Dataset sequence;
the model, checkpoint, software, or repository does not contain Raw Data;
the Result is not a Competing Dataset or prohibited motion-generation system;
the Result is not commercially distributed or deployed; and
the required attribution is included.

Publication of a Result does not grant any third party a licence to the Dataset.

11. FEEDBACK

Licensee may voluntarily provide corrections, suggestions, or feedback regarding the Dataset.

By providing feedback, Licensee grants Licensor a worldwide, perpetual, irrevocable, non-exclusive, royalty-free licence to use, reproduce, modify, and incorporate that feedback into Licensor’s datasets, products, services, and documentation without compensation or obligation.

12. DATASET CHANGES, SUPPORT, AND AVAILABILITY

Licensor may correct, update, replace, restrict, suspend, or discontinue the Dataset or any part of it at any time.

Licensor is not required to provide:

maintenance;
technical support;
corrections;
updates;
compatibility assistance; or
continued repository access.

13. DISCLAIMER OF WARRANTIES

THE DATASET IS PROVIDED “AS IS,” “AS AVAILABLE,” AND “WITH ALL FAULTS.”

TO THE MAXIMUM EXTENT PERMITTED BY APPLICABLE LAW, LICENSOR DISCLAIMS ALL EXPRESS, IMPLIED, AND STATUTORY WARRANTIES, INCLUDING WARRANTIES OF:

MERCHANTABILITY;
FITNESS FOR A PARTICULAR PURPOSE;
ACCURACY;
COMPLETENESS;
TITLE;
NON-INFRINGEMENT;
COMPATIBILITY;
SAFETY;
RELIABILITY; AND
ERROR-FREE OR UNINTERRUPTED OPERATION.

LICENSOR DOES NOT WARRANT THAT:

THE DATASET IS COMPLETE, ACCURATE, OR ERROR-FREE;
THE DATASET IS SUITABLE FOR A PARTICULAR MODEL, SOFTWARE PACKAGE, SKELETON, SIMULATOR, ROBOT, OR PIPELINE;
ANY MOTION IS PHYSICALLY SAFE OR ROBOTICALLY EXECUTABLE;
THE UNITREE G1 CSV FILES ARE DIRECTLY DEPLOYABLE ON A PHYSICAL ROBOT;
USE OF THE DATASET WILL PRODUCE A PARTICULAR RESULT; OR
ACCESS TO THE DATASET WILL REMAIN AVAILABLE.

Licensee is solely responsible for validation, testing, retargeting, simulation, balance analysis, collision checking, joint-limit checking, controller conversion, physical safety review, and legal compliance.

14. LIMITATION OF LIABILITY

TO THE MAXIMUM EXTENT PERMITTED BY APPLICABLE LAW, LICENSOR AND ITS PARTNERS, DIRECTORS, OFFICERS, EMPLOYEES, CONTRACTORS, AND AGENTS WILL NOT BE LIABLE FOR ANY:

INDIRECT;
INCIDENTAL;
SPECIAL;
EXEMPLARY;
PUNITIVE; OR
CONSEQUENTIAL

DAMAGES, OR FOR LOSS OF PROFITS, REVENUE, BUSINESS, GOODWILL, DATA, MODELS, OPPORTUNITY, OR EXPECTED SAVINGS, ARISING FROM OR RELATED TO THE DATASET, RESULTS, OR THIS AGREEMENT.

TO THE MAXIMUM EXTENT PERMITTED BY LAW, LICENSOR’S TOTAL AGGREGATE LIABILITY ARISING FROM OR RELATED TO THE DATASET OR THIS AGREEMENT WILL NOT EXCEED THE GREATER OF:

the amount paid by Licensee specifically for access to the Dataset during the twelve months immediately preceding the event giving rise to the claim; or
USD 100.

Nothing in this Agreement excludes liability that cannot lawfully be excluded.

15. INDEMNIFICATION

To the extent permitted by applicable law, Licensee agrees to indemnify, defend, and hold harmless Licensor and its partners, directors, officers, employees, contractors, and agents from claims, losses, liabilities, damages, penalties, costs, and reasonable legal fees arising from or related to:

Licensee’s access to or use of the Dataset;
Licensee’s use or distribution of Results;
Licensee’s breach of this Agreement;
Licensee’s violation of applicable law or third-party rights; or
any model, software, system, publication, service, or deployment created by Licensee using the Dataset.

16. SUSPENSION AND TERMINATION

16.1 Term

This Agreement begins when Licensee first accepts, accesses, downloads, copies, or uses the Dataset and continues until terminated.

16.2 Suspension

Licensor may suspend repository access while investigating suspected misuse or breach.

16.3 Termination for Breach

Licensor may terminate this Agreement immediately by written notice if Licensee breaches a material term.

16.4 Obligations After Termination

Upon termination, Licensee must:

immediately stop accessing and using the Dataset;
delete all copies of the Dataset and Raw Data under Licensee’s possession or control;
remove unauthorised public or third-party copies;
stop distributing or newly using Results that violate this Agreement; and
provide written confirmation of deletion within ten business days if requested by Licensor.

Deletion must include local copies, cloud copies, caches, repositories, shared storage, and backups where reasonably practicable.

Termination does not require withdrawal of an academic publication lawfully published before termination, provided it continues to comply with the attribution and non-disclosure requirements of this Agreement.

16.5 Survival

Sections concerning ownership, restrictions, attribution, warranties, liability, indemnification, termination, governing law, and provisions intended by their nature to survive will survive termination.

17. EXPORT CONTROL AND SANCTIONS

Licensee represents that Licensee is not prohibited from receiving or using the Dataset under applicable sanctions or export-control laws.

Licensee must not access, transfer, export, re-export, or use the Dataset in violation of applicable laws or restrictions.

Licensor may suspend or terminate access where Licensor reasonably believes continued access may violate such laws.

18. GOVERNING LAW AND JURISDICTION

This Agreement is governed by the laws of India, without regard to conflict-of-law principles.

Subject to any mandatory legal rights or procedures, the courts located in Hyderabad, Telangana, India will have exclusive jurisdiction over disputes arising from or related to this Agreement.

19. GENERAL TERMS

19.1 Entire Agreement

This Agreement constitutes the entire agreement between Licensor and Licensee concerning the Dataset under this Non-Commercial Research License.

A separate written commercial licence will control to the extent of any conflict regarding Commercial Use.

19.2 No Assignment

Licensee may not assign, transfer, delegate, or sublicense this Agreement or any right granted under it without Licensor’s prior written consent.

19.3 Severability

If any provision is held invalid or unenforceable, that provision will be modified only to the minimum extent necessary, and the remaining provisions will remain effective.

19.4 No Waiver

Failure to enforce a provision is not a waiver of that provision or any later breach.

19.5 Headings

Headings are for convenience only and do not affect interpretation.

19.6 Electronic Acceptance

Electronic acceptance, an access request, repository approval, download, API access, script access, or use of the Dataset may evidence acceptance of this Agreement to the extent permitted by applicable law.

19.7 Changes to this Agreement

Licensor may publish an updated version of this Agreement.

Material changes will apply to continued access, new downloads, or new use after the effective date of the updated version.

Licensor may provide notice through the Repository, the email address associated with an access request, or another reasonable method.

A Licensee that does not agree to an updated version must stop using the Dataset and comply with the deletion requirements in Section 16.

19.8 Notices

Notices to Licensor must be sent to:

Apple Arts Studios
Email: hello@appleartsstudios.com
Website: https://www.appleartsstudios.com/

SCHEDULE 1 — DATASET SPECIFICATIONS

1. Dataset Identity

Dataset Name: Apple Arts Studios Motion Dataset
Repository: AppleArtsStudios/Motion_Dataset
Producer and Licensor: Apple Arts Studios LLP, trading as Apple Arts Studios

2. Dataset Purpose

The Dataset contains professionally captured and processed human motion data intended for permitted academic research, AI research, robotics research, computer-vision research, motion analysis, and related Non-Commercial Uses.

3. Current Dataset Volume

The current release contains approximately:

199 h 30 min of Original captured motion;
full-length continuous data;

Mirrored sequences are augmented versions of Original captured motions and must not be described as separately captured performances.

4. Data Representations

4.0 SOMA Uniform BVH

Folder 02 Motive_Bvh
Format: .bvh
Frame Rate: 30 FPS

4.1 SOMA Uniform BVH

Folder: 03 Soma_Bvh
Format: .bvh
Frame Rate: 30 FPS
Description: Human motion retargeted to a uniform SOMA skeleton.

4.2 Unitree G1 Retargeted Motion

Folder: 04 Unitree_G1_Csv
Format: .csv
Description: Motion trajectories retargeted to a Unitree G1 representation.

The Unitree G1 CSV files are research representations and are not represented or warranted as direct commands for physical robot deployment.

Independent simulation, balance validation, collision checking, joint-limit validation, controller conversion, and physical safety testing are required.

5. Motion Variants

The Dataset includes:

Original motion sequences; and
Mirrored augmentation sequences.

6. Organisation and Documentation

The Dataset may include:

hierarchical motion categories and subcategories;
action and performance labels;
anonymised or pseudonymised actor identifiers;
motion identifiers and filenames;
technical or semantic descriptions;
skeletal metadata;
temporal metadata;
repository documentation; and
structured folder hierarchies.

The selection, arrangement, taxonomy, structure, and compilation of these materials form part of the Dataset and are governed by this Agreement.

SCHEDULE 2 — REQUIRED ATTRIBUTION

General Credit

Motion Data by Apple Arts Studios — https://www.appleartsstudios.com/

Academic Citation Text

This work uses the Apple Arts Studios Motion Dataset. Motion Data by Apple Arts Studios: https://www.appleartsstudios.com/

Model or Software Notice

Training data includes motion data from the Apple Arts Studios Motion Dataset. Use of the underlying Dataset is governed by the Apple Arts Studios Motion Dataset License Agreement.

CONTACT

For commercial licensing, permissions, access questions, suspected misuse, or legal notices:

Apple Arts Studios
Email: hello@appleartsstudios.com
Website: https://www.appleartsstudios.com/