✦ PIXELOKAY CREATOR STUDIO

Gesture Presenter

Make a presenter point to your content, wave, nod or emphasize a moment—exactly when it matters. Arrange gestures on a timeline and export your animation locally.

✓ Precise timeline✓ 2D rigged character✓ Free & private

THE WORKSPACE

AI Gesture Presenter 2D animation

Design a scene, schedule movement and export a presentation-ready animation.

Live sceneClick inside the scene to set the pointing target

Ready

Click to position the gesture target

00:00.000:12.0

Gesture timelineChoose a cue, adjust its timestamp, or scrub the playhead

Scene length (seconds)

Export & saveKeep editable cues or render the animation

Video resolution

Frame rate

Video export takes approximately real time, uses your browser's supported encoder, and may vary across devices. Uploaded videos play muted in this editor; their original audio is not included in exports. PNG exports capture the current frame.

Scene & character

Give the presenter a stage.

HeadlineDescriptionBackground

PNG/JPEG/WebP image or supported video. Max 50 MB. Local only.

Character color Skin tone Hair color Scale 100%X position 79%Y position 86%

Gesture editor

Schedule a precise movement.

Select a cue or add a gesture to edit it.

Gesture

Start (s)

Length (s)

Pointing hand

Target X (%)

Target Y (%)

For Point gestures, choose a hand and click the preview or set X/Y. Targets are relative to the scene.

Move selected cue

AI gesture planner Optional

Describe a cue sequence and review it before applying.

Presentation direction

Uses an optional on-device text model to plan only the six supported 2D gestures. It does not track body poses or automatically rig an uploaded person. No cues are changed until you approve.

Load a local AI model to generate a draft. Continue editing manually without AI.

How it works

Choose a scene and position the presenter.
Add gestures at specific times.
Preview by scrubbing or playing.
Download video or save an editable project.

What “AI” means here: Optional locally loaded AI can propose a sequence of six supported gestures from text. The character itself uses deterministic 2D rig animation—not pose estimation, arbitrary photo rigging, or freeform generated motion.

Put gestures exactly where your audience needs them.

☞

Targeted pointing

Direct the presenter's hand toward a particular place in your video or slide, using percentage coordinates.

◷

Frame-by-frame control

Schedule, preview, duplicate and adjust gesture cues along a continuous timeline.

◈

Private rendering

Animation and recording happen on your device. Video support depends on browser MediaRecorder capabilities.

Frequently asked questions

Can the tool animate a photo of a real person?

Not in this version. The built-in vector character is a deterministic rig; arbitrary images do not include arm and head bones.

Does the exported video include my video's original audio?

No. Uploaded videos play muted during preview, and the current video recorder captures the composed canvas silently. You can combine the exported animation with the original audio in a video editor.

Does the project JSON contain uploaded media?

No. To avoid huge project files, only the scene settings and gesture cue data are saved. Reattach your background video or image when reopening a project.

Can I export MP4?

When your browser advertises and successfully provides an MP4 MediaRecorder encoder, yes. Otherwise the tool uses WebM. Some browsers support only one format.

Using Gesture Presenter

Choose the scene and character, schedule non-overlapping gestures and review target positions before export.

Optional AI suggests cue schedules, but no generative motion, arbitrary photo rigging or lip sync. Source video audio is not exported. Projects omit media. Transparent PNG is a single frame, not transparent video. Scene length is bounded to 180 seconds; MediaRecorder frame timing and codecs vary.

Schedule procedural gestures and check their targets

Choose a character and scene, arrange the six supported gestures as sequential cues, and adjust the pointing target and timing. You can draft a cue schedule manually or opt into on-device AI planning; review every cue before exporting frames, a cue sheet, a project or recorded video.

Example workflow

Choose a character, background and scene length, then add non-overlapping gesture cues in the intended order.

Set the pointing target where needed, review the cue schedule and adjust positions or timing by hand.

Export a PNG for one frame, a transparent rig PNG, CSV/JSON cue data or a browser-recorded video.

Authored example: Build a three-cue product walkthrough with a wave, a point toward a labeled feature and a closing gesture. Inspect the target position and spacing between cues before recording.

Limits and model handling: Optional WebLLM planning requires explicit model setup, compatible WebGPU and external runtime/model downloads; the manual gesture editor and exports work without it. AI drafts cue schedules rather than generating motion. Source video audio is not exported, saved projects omit background media, and transparent PNG represents one frame rather than transparent video. Scene length is capped at 180 seconds; recording timing and codecs vary by browser.

Gesture Presenter questions

Do I need AI to edit gestures or export a video?

No. Manual cues, scene controls and exports work without loading a model. AI is an optional cue-planning aid.

Does the presenter rig lip-sync to speech or animate arbitrary photos?

No. Gestures are procedural and do not provide lip sync or arbitrary photo rigging.

Gesture Presenter

AI Gesture Presenter 2D animation

Put gestures exactly where your audience needs them.

Targeted pointing

Frame-by-frame control

Private rendering

Frequently asked questions

Processing

Using Gesture Presenter

Schedule procedural gestures and check their targets

Example workflow

Gesture Presenter questions

Related tools

Set up AI on this device

Device & model resources

Free AI model storage across PixelOkay

Delete recognized AI model downloads?