Everything you show — a page of your slides, an application, a drawing, your webcam — sits on one picture, one thing on top of another. SplitCam calls each of them a layer, and it hands the finished picture to your meeting app as an ordinary camera: Zoom, Google Meet, Microsoft Teams and 60+ apps in total find it in the camera dropdown. Your class sees you and the material at the same time, and the share-screen, unshare, "can everyone see this?" routine stops being part of the lesson.
These are the layers worth knowing about for teaching. Several of them exist on one platform only, so the platform is named on every card — take the label seriously, because the worst time to discover a missing source is mid-lesson.
SplitCam goes on the computer you already teach from — Windows 10 or 11, or macOS 13 or later — and adds one more camera to the ones your system already offers. The half of this that matters is the half that does not change: your students open the same meeting link, in the same app, on the same device, and are never told that anything is different. If the teaching machine is administered by someone else, they will need to run the installer for you.
The lesson already exists somewhere on your disk: the presentation you exported, the worksheet you scanned, the site you meant to demonstrate, the folder of photographs. Drag the file straight onto the scene and it becomes a layer. What changes teaching is that they can all be live at once rather than one at a time — you stop choosing which single thing the class is allowed to see. Add them once and they are still there next term. The cards further down cover each type and name the platform it is on. On Windows, add an Application Audio source too if you plan to play a clip, so the clip's sound reaches the class alongside your microphone.
Drag the material until it fills most of the frame, then put your webcam in a corner you are not going to write in — bottom right, for most people. Now judge it the way your class will: whatever you show arrives at the other end inside a tile roughly the size of a postcard, on a laptop or on a phone held at arm's length. If you cannot read your own axis labels at that size, nobody can. On Windows you can also magnify the whole picture — point at the detail and roll the mouse wheel with no layer selected. Click the Lock icon beside your webcam in the media list first and it keeps its place while everything else grows.
Arrange the layers once, then save the arrangement as a scene in the scene list, and add one scene per thing you actually do — a lecture view, a worked-example view, a demonstration view. Switching is a click on the scene you want; on Windows you can also put each scene on a hotkey and change layouts without reaching for the mouse. What matters is how it looks from a student's seat: the material changes, the picture stays the same size and shape it was a second ago, and the camera never cuts out. Nobody sees a black tile, and you never say "let me stop sharing" again.
Open the camera setting in your meeting app and pick "SplitCam Camera" from the list your webcam is already in — in Zoom and Google Meet it sits under Settings → Video, in Microsoft Teams under Settings → Devices. It arrives as an ordinary webcam in 60+ apps, with no plugin and no per-app integration, so the same list works mid-lesson: switch to it after the class has started and the picture simply changes over. On macOS the camera is a system extension, so the Mac asks you to allow it the first time.
Export your slides to PDF, or take the worksheet you already have, and add the file as a PDF layer. You turn pages inside the layer while the lesson runs, so a question about page four is answered on page four instead of after a hunt through a menu.
Add a Slideshow layer and put your images in it: reorder them, remove the ones you did not need, advance them on a fixed interval, loop the set, and use an animated GIF where a few seconds of motion explain more than a still. On macOS you can also drag three or more images straight onto the scene and the Slideshow layer builds itself.
A layer you draw on free-hand with the mouse or trackpad; hold Shift and the line comes out straight, and the layer has Undo, Redo and Clear All. Free-hand is all it does, which makes it a place to work a problem through in front of the class rather than a way to reproduce a printed figure. The macOS build does not have it.
Capture the whole screen, a single application window, or just the part of the screen that matters, up to 4K, with the cursor visible so the class can follow where you are pointing. On Windows, Full Screen capture can also show your mouse clicks and key presses — the difference between a software walkthrough students can follow and one they can only watch.
Add a Browser layer for anything that lives in a browser: a web-based slide tool, a graphing site, a simulation. Double-click the layer to enter interaction mode and you can click and type inside the page while the class is watching it, without leaving SplitCam or opening another window.
Add a Phone Camera source and scan the QR code with your phone. It joins over the network, with no cable and no capture hardware, so you can point it at a book, a printed diagram or an object on the desk and switch to that angle when the explanation needs a real thing in it.
A text layer holds a title, an instruction or the key term you keep repeating, with its own colour, a background panel behind it, and an outline round the letters so it stays readable over a busy slide. On Windows it can also translate what you type into another language automatically, which is worth knowing if part of your class follows in a second language. It translates typed text, not what you say out loud.
Legibility is the whole job here. Every arrangement below is decided by one question — what is the class meant to be reading right now, and is it big enough to read? Build them once, save each as a scene, and the rest of the term is a key press.
A PDF page or a Slideshow image takes almost the whole picture; your webcam sits small in a corner, clear of the text. The class reads the slide at close to the size a full-screen share would have given them, and still sees who is speaking and where you are looking.
On Windows, a Whiteboard layer takes the working area and you write into it as you talk. Write two or three times larger than feels natural — you are writing for a tile, not for your monitor — and clear the board between problems rather than scrolling for fresh space.
Window Capture of the software you are teaching, with the cursor visible and, on Windows, your clicks and key presses shown as well; your webcam stays small in its corner. The class sees which menu you opened and which key you pressed, not only the result that appeared.
One scene that is nothing but your webcam, full size. Switch to it when you stop showing and start asking, and the class gets your face at a size where expression reads. It is also the scene to have ready when a student asks something the slides do not answer.
Before the lesson, shrink the SplitCam preview until it is about the size of the tile a student actually gets, and look at your own material from across the desk. Anything you have to lean in to read is too small — bigger type, fewer things on screen at once, and one idea per arrangement rather than a full page shrunk to fit.
Pick the corner your webcam lives in and then never write, point or drop a diagram into it. A layer you have to move mid-lesson is a layer the whole class watches you move.
Arrange the scene so the thing you are talking about sits high on your screen, near the camera. You end up looking at the class instead of down and off to one side, and nobody spends the lesson wondering what you are really reading.
For a fixed-length activity, an Automation Timer layer counts down in view of the class, so nobody has to ask how long is left. On Windows the timer can switch scenes when it runs out, so the lesson comes back on its own instead of when you remember it.
Five minutes of changing scenes with nobody watching is enough. The lesson you already know; the only new muscle memory is which scene holds which arrangement — and on Windows, which key brings it up. Do it once and you stop thinking about the software during the lesson, which is the entire point of setting it up.
SplitCam writes the picture the class is seeing to a file on your own machine, so a student who missed the session gets exactly what everyone else saw. On Windows the file is written as fragmented MP4, so a crash or a power cut does not cost you the whole recording.
Not a vendor table — these are the four things teachers already do to show something in a lesson. Each of them does something SplitCam does not, and SplitCam's part is narrower than a comparison chart would suggest. The meeting app still runs the class.
It works and every meeting app has it. But your face becomes a thumbnail beside the share rather than part of the same picture, every change of material means finding another window and re-picking what to share, and anything that is not open on your computer — a book, a sketch you just made — cannot go in at all.
Nothing else shows your hands working on a real page, and for physical material it is still the best answer. It is one fixed angle, though, and one more device to own. It is also not an either-or: most connect as an ordinary webcam, so SplitCam can take it as one layer beside your slides rather than instead of them.
The handwriting is good and the pen is the right tool for it. It needs a second device, a mirroring app and a share window, and while the mirror is up your face is a thumbnail next to the share rather than part of the same picture the class is reading.
Ink drawn on your slides inside the presentation app looks right and stays with the slide. It only exists inside that app, though: move to a browser or a piece of software you are demonstrating and the pen has nothing to write on.
SplitCam assembles the picture and hands it to the meeting app as a camera. You and the material stay in one frame, arrangements change without anything stopping, and the material can be a PDF, a screen, a web page or your phone's camera at the same moment. The drawing is free-hand with a mouse or trackpad, on Windows.
No. SplitCam runs on your computer only, where it appears as an extra camera. Your class joins the same meeting link in the same app they always use, and what reaches them is a camera picture — the same kind of video their own webcams send.
Yes to both. Open Settings → Video in Zoom, Teams or Meet and "SplitCam Camera" is in the camera dropdown alongside your webcam; it shows up the same way in 60+ apps, with no plugin and no per-app integration. You can open that dropdown after the lesson has started, and the picture changes over without leaving or rejoining.
Two honest routes. Export the slides to PDF and add the file as a PDF layer — then you turn pages inside the layer while you teach. Or, if you would rather drive PowerPoint or Keynote directly with all its animations and its own pen, use Window Capture on the presentation window and put that in the scene instead.
What SplitCam documents is a Whiteboard layer of its own, on Windows: free-hand paths, Shift for a straight line, Undo, Redo and Clear All. It is a surface to draw on, not a marker that floats over your other material — so the usual arrangement is the slide taking most of the frame with the board beside it, or a scene you switch to when the working starts. If you want ink that sits on the slide itself, draw with PowerPoint or Keynote's own pen and capture that window.
The whiteboard is documented as free-hand drawing with a mouse or trackpad; there is nothing published about pen pressure or stylus support, so do not plan a lesson around it. If pen input is central to how you teach, draw in an application that supports your tablet properly and bring that window into the scene with Window Capture.
Add the clip as a video layer and it plays inside the scene. For sound coming out of another program, such as a video in a browser tab, Windows has an Application Audio source that takes that program's sound directly and mixes it with your microphone. The microphone's Accentuate option quietens the other sources whenever it hears you, so you can talk over a clip — it reacts to any sound rather than only your voice, so wear a headset when you use it.
Yes. SplitCam records the assembled picture to a file on your own machine — no cloud, no portal, no link, so you share the file however your school already shares files. On Windows it is written as fragmented MP4, which means an interrupted recording is still a playable one.
The core does: PDF layers, Slideshow, screen and window capture, the Browser layer, Phone Camera, recording and the virtual camera itself all run on macOS 13 or later, where the camera is installed as a system extension you allow on first launch. Documented on Windows only: the Whiteboard, Application Audio, hotkeys, the Replay source and the solid Color source. If your teaching leans on the whiteboard, teach from Windows.
Take a lesson you have already prepared, spend a few minutes putting its material into a scene, and teach it. None of it lands on your class as extra work: same meeting link, same app, nothing to install, no minute spent explaining a new tool. The only difference they can see is that you and the thing you are explaining are finally in the same picture — and if that turns out not to suit the way you teach, you pick your own webcam back out of the same dropdown and the lesson carries on.
⬇ Free Download