God's Eye View Live 3D Globe Released

A photorealistic 3D globe built only from public signals is attracting developers’ attention. According to coverage of bilawalsidhu on GitHub Trending, aircraft, ships, satellites, earthquakes, traffic, and public camera feeds can be overlaid and explored on a single screen. Operation is completed entirely in the browser, with voice control by a real-time AI agent. The repository ranked No. 1 on GitHub Trending daily and weekly in August 2026. Public information says it placed 8th for Product of the Day.

The origin is the God’s Eye View series that became a topic on video-sharing platforms. Formerly called WorldView, it gained over 5 million views on YouTube and over 25 million views across social media as a whole. This release organizes that project as reproducible code. The match between the creator’s background and measured data is seen as having pushed interest higher.

A spy-satellite simulator in your browser — then you realize the sources are public and the data is real.

The above is a quotation from the repository’s introduction. Relying on fair quotation under Article 32 of the Japanese Copyright Act, the original expression is shown verbatim. Its appearance resembles a reconnaissance satellite console. But the claim is that the references are public materials and the data itself is real.

A 3D Globe Binding Real Data Together

The core of this work is the ability to move back and forth between a global overview and details of individual objects. Aircraft transponder signals, ship identification signals, orbital elements, seismometers, and public camera feeds are the source materials. Users rotate the globe, select an object of interest, and zoom in. The move from overall awareness to individual tracking connects smoothly.

Rendering emphasizes a photorealistic texture. Terrain relief and day-night shading function as a background for situational awareness. The layer structure is separated, allowing aviation, maritime, satellite, earthquake, traffic, and video layers to be overlaid. Each layer can be added or replaced as an independent component.

Processing runs on the user’s computer. The code is inspectable and extensible, and is designed not to depend on opaque external processing. Most sources are live or regularly updated. However, traffic is simulated along real roads, based on aggregated location data. Note that surveillance camera positions and launch trajectories are rough estimates.

Full Overview of the Operation System for

Tracking and Onboard Views

The tracking function is based on a locked view of the selected object. When an object is selected, the camera follows it, and its movement trail remains while fading. Related specifications are displayed together to aid situational awareness. When tracking fires or ships, you can move to the nearest live footage in one step.

Cockpit view is a function to ride inside the aircraft being tracked. Terrain flows by below, maintained continuously through descent. It intuitively conveys the sense of speed and altitude of movement. It is also easy to use as material for recorded video.

It also includes a directory function for the surrounding 250 km. It lists aircraft and related objects near the target, allowing sequential switching. If you select an aircraft of interest, you can move directly to its cockpit view. It suits uses that move back and forth between wide-area monitoring and individual inspection.

As shown in NVIDIA Jensen Huang Visits Japan, Three Transformations for Japan, GPU performance is the foundation of photorealistic expression. This work is also expected to have a significant rendering load. Checking computer performance before use is essential.

Implementation of Voice Control and Video

Presentation Features

Hands-free voice control is a feature of this work. A real-time AI agent interprets speech and converts it into globe operations. You can move and switch layers without using your hands. It is suitable for spoken instructions during work and operation during demonstrations.

Voice whiteboard is a function to write annotations into the world by speech. It can draw actual boundary polygons, symbols, and routes. Spoken instructions can be turned into visible drawings. It is useful for sharing deliberation processes and reviewing afterward.

Aircraft representation also reflects real types. 3D models are prepared for each type, such as the 787, ATR-72, Citation, Bell 206, and MQ-9. They are displayed as symbols at a distance, switching to 3D models on approach. This is a device to reduce identification load while increasing concreteness when close.

Features to change the look of the video are also extensive. As GLSL-based instrument-style looks, CRT, NVG, FLIR/thermal, Noir, and Snow can be selected. There is also a detection display that overlays boxes and identifiers on objects in the screen. The tactical-style HUD overlays measurements such as speed and altitude.

The director function records and replays camera movement paths. It is suitable for creating short introductory videos and demonstration materials. There is also a wide-area context function that arranges the overall situation in one step. After finishing, you can return to the original viewpoint.

Shareable links support reproducibility. Camera, picture tone, layer composition, and even tracked objects can be encoded into a URL. A URL containing live objects functions not just as a screen but as a handoff. As with the conversational assistant shown in Instacart Unveils Conversational AI Assistant Clementine, the combination of spoken instructions and state sharing is key.

Operating Environment and Optional Key

Management Explained

It is designed to get started without keys. It supports installation via Pinokio or local execution from the terminal. Optional keys can be added inside the app. The ability to try it while keeping initial costs down is commendable.

The fact that each layer is a separate component is important for developers. Adding a source is treated as adding a layer. It is easy to connect proprietary data and change display formats. Combined with branch management as in Git Flow vs GitHub Flow vs Trunk Based Development: 2026 Branch Strategy Comparison, collaborative development is easier.

The cost structure depends on user choices. Basic operation can start with the included data sources. Using external live sources or voice agents may incur costs depending on keys. You need to check the conditions for each key in advance.

Organizing Notes on Openness and Accuracy

The value of this work lies in the integration of public signals. Flight transponders, ship beacons, orbital elements, and seismometers are highly open. Placing them on the same platform makes comparison and contextualization easier. Use in education, reporting, and organizing situations during disasters is conceivable.

On the other hand, variations in accuracy are explicitly stated. Traffic is not measured, but simulated along real roads. Surveillance camera positions and launch trajectories are rough estimates. Readers are required to distinguish live from simulation, and measured from estimated.

Handling of public video also requires care. Even though public, use leading to identification of individuals or vehicles should be avoided. When a shared URL contains live objects, managing the scope of dissemination is necessary. Use under established operational guidelines is desirable.

Inspectability supports trust. Because the code can be checked directly, the basis for displays can be traced. Update intervals and handling of missing data for each source can also be confirmed. As a verification item at introduction, listing data sources is effective.

Editorial Opinion

In the short term, we expect derivative development starting from public-signal visualization to increase. The per-layer separation lowers barriers to adding features. We assess that use via video effects and shareable links will spread in demonstrations and verification settings.

In the long term, we see the boundary between live awareness and recording becoming a challenge. Integration of public information increases convenience while also raising the risk of misreading. Methods for distinguishing simulation from measurement and for provenance management will grow in importance.

The remaining issue is the reliability and reproducibility of voice operation. Real-time agent responses are affected by ambient noise and phrasing. The question is how far the sharing format for operation records can be standardized.

References

• “bilawalsidhu / gods-eye-view”, by bilawalsidhu — GitHub Trending, 2026-09-13 (ARR)
• Source URL: https://github.com/bilawalsidhu/gods-eye-view