VOLE-Camera — Sensor-native procedural entropy factorization and direct visual materialization.
Find a file
2026-09-05 04:22:47 +01:00
assets Add VOLE-Camera logo at 25% size and RAW10 boundary adapters 2026-09-05 03:11:14 +01:00
docs Phase A: repository foundation and canonical sensor-domain types 2026-09-05 03:09:12 +01:00
src Add VOLE-Camera logo at 25% size and RAW10 boundary adapters 2026-09-05 03:11:14 +01:00
.gitignore Phase A: repository foundation and canonical sensor-domain types 2026-09-05 03:09:12 +01:00
Cargo.lock Bump version to 0.1.1 2026-09-05 04:22:47 +01:00
Cargo.toml Bump version to 0.1.1 2026-09-05 04:22:47 +01:00
CONFORMANCE.md Phase A: repository foundation and canonical sensor-domain types 2026-09-05 03:09:12 +01:00
LICENSE-APACHE Add MIT and Apache-2.0 license texts 2026-09-05 03:12:21 +01:00
LICENSE-MIT Add MIT and Apache-2.0 license texts 2026-09-05 03:12:21 +01:00
PROJECT_STATE.md Phase A: repository foundation and canonical sensor-domain types 2026-09-05 03:09:12 +01:00
README.md Add prior-art citation under the logo in README 2026-09-05 04:22:14 +01:00
rust-toolchain.toml Phase A: repository foundation and canonical sensor-domain types 2026-09-05 03:09:12 +01:00
SECURITY.md Phase A: repository foundation and canonical sensor-domain types 2026-09-05 03:09:12 +01:00

VOLE-Camera

Sensor-native procedural entropy factorization and direct visual materialization.

VOLE-Camera logo

de Beer, R. (2026). VOLE-Camera: Sensor-Native Procedural Entropy Factorization and Direct Visual Materialization - Broad Prior-Art Technical Disclosure and Research Architecture (Version v1.0). Zenodo. https://doi.org/10.5281/zenodo.22312959

VOLE-Camera is an experimental native-Rust implementation of sensor-native procedural entropy factorization. Its current proof is a deterministic simulated sensor pipeline; physical-camera claims remain unsealed.

VOLE-Camera is a separate research repository from infinityabundance/vole (VOLE — Video Object Layer Engine), but conceptually downstream of it. It owns the camera-specific layer: the sensor sample domain, capture state, rolling-shutter timing, camera telemetry, calibration state, sensor-space predictors, a camera-native root seed, and camera-specific evidence. It does not fork VOLE and does not duplicate its general visual inverse engine.

The research question

How much of a captured sensor stream can be represented economically as persistent deterministic capture state, camera telemetry, reusable visual structure, bounded procedural evolution, and exact predictive models before the residual information dominates? And does starting the representation at the camera preserve useful deterministic state that is lost when sensor observations are first flattened into conventional raster/video frames?

The system is built to be falsifiable, not to demonstrate favorable synthetic examples.

Status

Current head: Phase A / B in progress (see PROJECT_STATE.md). No hardware code exists yet; it may not begin before the simulator seal (§239). The first complete deliverable is vole-camera demo, which runs entirely on an ordinary Linux desktop without a camera (§13).

Prior art

See docs/reread-sources.md for the pinned versions of the sibling research repositories and the prior-art paper (de Beer, 2026, DOI 10.5281/zenodo.22312959). docs/non-claims.md is the normative statement of what this project may and may not claim.

License

Dual-licensed under the MIT or Apache-2.0 terms, at your option.