Bionic Eye
An fun experience
$19.99 · Designed for iPad. Not verified for macOS.
Learn about the technology of the bionic eye
Experience looking through a bionic eye.
This app turns a live camera feed into a low-resolution phosphene display that simulates how a visual implant might present vision. It rotates the image, extracts brightness and edges, compresses the scene into an electrode grid, and maps each cell to a brightness value that becomes the on-screen output.
The main challenge is reducing a detailed camera image without losing important shapes, boundaries, and contrast. To solve that, the pipeline uses luminance extraction, edge detection, grid sampling, and spike-rate encoding before rendering the final phosphene map.
The transformation process begins with a rotated camera frame that is converted into a uniform raster buffer, then passed through optical degradation, luminance extraction, center-surround filtering, and orientation-sensitive edge reinforcement. Each stage compresses the scene while preserving high-information structure: edges, discontinuities, and local contrast. The result is a retina-cortex surrogate that resembles a sparse biological encoding rather than raw pixels.
From there, the image is projected onto an electrode lattice. Each electrode corresponds to a spatial region of the scene, so the grid is a many-to-one mapping from camera space into implant space. The **spike grid** is the intermediate neural representation: a 2D array of firing-rate values, one per electrode, where magnitude encodes how strongly that region should stimulate the visual system. It is not color and not display brightness directly; it is a surrogate for neural activation strength after retinal and cortical preprocessing.
The **single electrode** is the atomic output unit of the model. For one cell, the pipeline takes its local spike rate, normalizes it into a bounded activation domain, modulates it with temporal pulse phase, and combines it with a per-electrode color sample from the separate color path. That yields the final phosphene element: a point whose opacity, chroma, and apparent intensity approximate what one implant contact might evoke perceptually.
Technically, the display is therefore a spatiotemporal decoding layer over the spike grid. The grid carries neural state; the electrode cell converts that state into rendered light. This separation is important because it lets the simulation preserve biological fidelity in the upstream encoding while still producing a legible, interactive phosphene visualization downstream.
Ratings & Reviews
- This app hasn’t received enough ratings or reviews to display an overview.
The developer, David Nishimoto, indicated that the app’s privacy practices may include handling of data as described below. For more information, see the developer’s privacy policy .
Data Not Collected
The developer does not collect any data from this app.
Accessibility
The developer has not yet indicated which accessibility features this app supports. Learn More
Information
- Seller
- David Nishimoto
- Size
- 7.5 MB
- Category
- Business
- Compatibility
Requires iOS 26.0 or later.
- iPhone
Requires iOS 26.0 or later. - iPad
Requires iPadOS 26.0 or later. - Mac
Requires macOS 26.0 or later and a Mac with Apple M1 chip or later. - Apple Vision
Requires visionOS 26.0 or later.
- iPhone
- Languages
- English
- Age Rating
4+
- 4+
- Copyright
- © 2026
