How It Works
What does VisualAim actually do?
VisualAim is a real-time computer vision tool that captures your screen, runs a YOLOv8 detection model against the image, and can trigger configurable input responses based on what it detects. The overlay renders detection results directly onto your display. All processing happens locally on your Windows machine — no cloud, no external hardware.
What can it detect?
The default model is trained on a general set of visual targets. For scene-specific detection, custom ONNX models can be loaded — detection quality scales directly with the model used. The detection region, confidence threshold, and FOV are all configurable.
Setup & Compatibility
What are the system requirements?
VisualAim requires Windows 10 or Windows 11. It supports all major GPU brands — Nvidia, AMD, and Intel. Any modern Intel or AMD CPU will work. There is no specific minimum GPU requirement as it supports all GPUs via DirectML.
What can VisualAim detect on screen?
VisualAim works with any visual content that can be detected by its underlying ONNX model. The default model covers a general set of targets. For specific use cases, you can load your own ONNX model to detect whatever objects or patterns you need.
Does it support controller input?
Yes. VisualAim can detect and respond to controller inputs. Note that VisualAim runs on Windows only — it does not run natively on console hardware like PS5 or Xbox.
Does it work with streamed or remote desktop content?
Yes. Since VisualAim reads your local display and does not interact with the source application, it works with any content visible on your screen — local or streamed.
Do I need any extra hardware like a KMBox or Arduino?
No. VisualAim requires no additional hardware whatsoever. No KMBox, Arduino, or any other external device is needed — it runs entirely on your existing PC using your standard mouse and keyboard.
My antivirus is flagging the software. Is it safe?
This is a false positive. VisualAim uses low-level Windows APIs for screen capture and input that antivirus heuristics can flag — this is common with any software using similar system-level access. The software does not contain malicious code. Add an exception in your antivirus before running.
Billing & Refunds
What payment methods do you accept?
Payments are processed through Polar Checkout. Available payment methods are shown directly in checkout before purchase.
Can I get a refund?
Refunds are available within 14 days of purchase if the software cannot run on your Windows 10 or 11 system. Refunds are not available after successful activation and use. Please verify compatibility before purchasing, as Windows 10 or 11 is required.
What does "Lifetime" mean — is it truly lifetime?
Yes. A Lifetime license gives you permanent access to VisualAim and all future updates at no extra cost. You pay once and never again.
Features & Configuration
What input response modes are available?
VisualAim supports two input response modes: tracking mode, which moves the cursor toward a detected target, and click response mode, which triggers a click when the cursor is within the detection zone. Both can be used together or independently and are configured via keybinds in the menu.
How do motion smoothing and prediction work?
VisualAim includes smoothing and prediction settings that shape how the cursor moves toward a detected target — configurable curve, reaction delay, and dynamic speed. These are fully tunable so you can match the movement profile to your use case.
Can I customize the settings?
Yes. VisualAim has a full web-based configuration menu where you can adjust detection zone, response strength, smoothing, click response delay, keybinds, overlay style, detection sensitivity, and more. Settings are saved per profile so you can maintain separate configurations for different detection scenarios.
Note: VisualAim is a computer vision and configurable input tool for Windows. It is your responsibility to ensure your use of this software complies with the terms of service of any application you use it with. We are not responsible for consequences arising from use in contexts that prohibit automated input.