Process a folder of tennis serve videos

Coaches can do this entirely in the browser. There is no script to install or command to type.

1. Open Process a folder from AI Tennis Serve Analysis in desktop Chrome or Edge.
2. Choose a folder and allow the browser to read and write it. Each video should hold one serve, filmed from a true side view with the full server and the ball in frame through contact.
3. Confirm capture FPS and serve type (flat, slice or kick). Override FPS for individual videos if needed.
4. Press Process videos and keep this tab open. The whole clip is analyzed, as in the single-video app.
5. Find the MP4 highlight, PDF report and Markdown text beside each original. Review each row for warnings or failures.

What gets saved

For serve.mp4, the app creates results_serve.mp4, results_serve.pdf and results_serve.md. A results_serve.json receipt records the settings, actual frame rates, analyzed range and output sizes. The MP4 is a metric highlight reel, with 30 FPS output.

Originals are not changed. Existing result files are skipped, including incomplete sets. Move old or partial results before retrying. Videos with the same base name, such as serve.mp4 and serve.mov, must be renamed first. Files beginning results_ or reslts_ are excluded. Subfolders are not processed.

Recording settings that matter

• Capture FPS: Auto reads the encoded file rate. Slow motion filmed at 120 FPS may be saved at 30 FPS; enter 120 in that case. Contact timing against the peak of the jump uses this rate. Auto fails if the file rate cannot be read, so you can choose a rate explicitly.
• Serve type: sets the toss-placement target the serve is scored against. Use separate batches for different serve types.
• Serve phases: toss, trophy, contact and landing are detected automatically, and the hitting hand is read from the video. A video in which no complete serve is found is marked failed with advice; the reports need a measured serve.
• Models: Thunder pose and the fastest ball/racket detector are the defaults, as in the single-video app. Larger detectors are slower; if one cannot load, the nano detector is used and the row shows a warning.

Local processing and recovery

Video analysis and export happen in your browser. Static app code and model files may download from the internet. Folder permissions are granted by you in the browser, and you may need to grant them again after reopening the page.

One video runs at a time. A failure does not stop the rest. Stop batch cancels the active analysis/export; a save already in progress finishes before the queue stops. Closing the browser can leave partial results. Move these before retrying. Avoid concurrent programs writing to the same folder.

If the folder picker is unavailable, use a supported desktop browser such as Chrome or Edge, or return to the single-video app. MP4 export requires H.264 WebCodecs support; an unsupported encoder produces an error. Results are the same estimates and coaching guidance offered by the individual-video workflow.

const api = window.apoptAutomation;
api.capabilities();   // app "tennis", views, options, defaults, outputs
const id = api.start({
  captureFps: 120, serveType: "kick", detectionModelSize: "s"
});
api.getJob(id);       // state: running → succeeded / failed / cancelled
// After state === "succeeded":
await api.markdown(id);   // the text results
api.download(id, "mp4");  // also "pdf" and "md"
api.cancel(id);           // stops a running job