MOCAP API
Use the AIMoCap video mocap API to create async jobs, upload source video, poll status, and download FBX or robot motion outputs.
For developers looking for an API-driven video mocap pipeline.
AIMoCap's mocap API is an async job lifecycle for uploaded clips, target IDs, v-credit accounting, polling, and artifact downloads across Default or Unitree G1 outputs.
Use the mocap API when motion processing must fit into a product workflow with upload URLs, target selection, status polling, result downloads, and API v-credit accounting.
Do not use the API when you need synchronous frame-by-frame processing or a private custom target that has not been prepared for your account.
The AIMoCap API is designed for production integrations that need asynchronous video mocap jobs instead of manual browser-only workflows.
Developers can create a job, upload video to the returned upload URL, complete the upload, poll job status, and download available outputs.
This makes the API page different from a generic product page: it should help an engineer understand job state, supported targets, billing units, and failure boundaries before writing client code.
The API is a fit when your product can manage an asynchronous job lifecycle instead of waiting for a single blocking conversion request.
Developer discussions about video-processing APIs usually focus on job lifecycle, upload reliability, billing predictability, retry behavior, and whether examples match the documented API.
API users expect long video jobs to be non-blocking, so create/upload/complete/poll/download should stay visible across docs, examples, and landing pages.
A recurring developer concern is whether usage cost can be estimated and audited. AIMoCap should keep v-credit separate from web credits and explain when the API ledger changes.
Integrators need to know what happens for invalid FPS, trim ranges, large files, insufficient v-credit, and processing failures before they build a product workflow.
Use these facts to decide whether this workflow matches your output, integration, and cleanup needs.
The public API is designed around create, upload, complete-upload, poll status, and download, so long-running mocap jobs do not block client requests.
A single API job can request supported targets such as Default and Unitree G1 when the account and job limits allow it; robot-target output should still pass downstream simulation, controller, or safety validation before hardware use.
API jobs use API v-credit, while browser Studio jobs use web credits; this keeps product integrations and manual use easier to audit.
Client code should be prepared to download different result artifacts depending on requested targets, such as FBX for Default output or robot-oriented data for Unitree G1.
A robust client stores job ID, requested target IDs, upload completion state, polling status, and downloaded artifact keys separately instead of treating the API as a single blocking conversion call.
Send job metadata such as title, supported target IDs, optional trim fields, and export FPS.
Upload the source video to the provided upload URL, then call complete-upload so AIMoCap can queue processing.
Poll the job status and download available FBX, preview video, or Unitree G1 robot motion output when complete.
Track API v-credit separately from browser Studio credits so product usage, automated jobs, and manual Studio jobs can be audited independently.
Yes. The API supports multiple target IDs in one job when the account and job limits allow it. Unitree G1 output is robot-oriented motion data, not a direct hardware-command stream.
API jobs use API v-credit after the uploaded source is validated for processing. Cost is based on the resolved processing duration, rounded for billing, and stays separate from web Studio credits. If AIMoCap has a system-side processing failure after charging, the API ledger should record the refund path instead of mixing it with Studio credit usage.
The API reference documents supported targets, trim fields, export FPS, upload, polling, and result download behavior.
No. AIMoCap API jobs are asynchronous: create the job, upload the video, complete the upload, poll status, and download results when processing finishes.
No. API jobs use API v-credit, while browser Studio mocap jobs use web credits. This separation keeps automated product usage easier to track.
Persist the job ID, requested target IDs, upload completion state, current polling status, and the result artifact URLs or downloaded files for each completed target.
Continue through this topic cluster to compare output formats, API options, and workflow boundaries.
Use output, limits, and accounting to evaluate fit.
Compare animation, preview, and robot artifacts.
Build async video mocap jobs with upload, polling, and downloadable motion outputs through the AIMoCap API.
Use AIMoCap API jobs to request Default output and download animation-ready FBX motion after processing.
Use target-aware AIMoCap API jobs when your integration needs robot motion output such as Unitree G1.
Create AIMoCap video-to-FBX API jobs, upload source clips, poll status, and download outputs when processing completes.
These related AIMoCap resources document the workflow boundaries, output formats, and implementation details referenced on this page.
