Expand description
cog-pose-estimation library surface.
See ADR-101 for the design and ADR-100 for the surrounding Cog
packaging spec. This crate is intentionally a thin shell around
wifi-densepose-train’s exported model types — the heavy lifting
(encoder, pose head) lives there.
Modules§
- config
- Runtime configuration for the pose-estimation Cog.
- inference
- Inference engine — loads
pose_v1.safetensors(produced by the Candle training run onruvultra’s RTX 5080, seecog/artifacts/pose_v1.safetensors+docs/benchmarks/pose-estimation-cog.md) and runs the encoder + pose head on each CSI window. - manifest
- Cog manifest — see ADR-100 §“manifest.json schema”.
- publisher
- Structured JSON event publisher — one line per event on stdout.
- runtime
- Long-running inference loop. Polls the appliance’s sensing-server,
runs a CSI window through the engine, emits
pose.frameevents.
Constants§
- COG_ID
- Cog identifier — matches the on-disk path
/var/lib/cognitum/apps/pose-estimation/. - COG_
VERSION - Cog version (sourced from Cargo.toml at build time).