berkus
/
wifi-densepose
mirror of https://github.com/ruvnet/wifi-densepose.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

docs(changelog): WiFi-CSI efficiency frontier + per-room calibration service 

Document the beyond-SOTA efficiency frontier (75K params beats SOTA, int4
edge model 20KB@74%), few-shot calibration resolving generalization
(cross-env 10->73%), and the calibration service (Python ref + Rust
cog-pose --adapter integration).

Co-Authored-By: claude-flow <ruv@ruv.net>
This commit is contained in:
ruv 2026-05-31 02:38:07 -04:00
parent 83299b4d04
commit 17ff2433bc
1 changed files with 4 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
CHANGELOG.md
View File

@ -8,6 +8,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
## [Unreleased] ## [Unreleased]
### Added ### Added
- **WiFi-CSI pose: efficiency frontier + per-room calibration service** (ADR-150 §3.2–3.6). Two beyond-SOTA results on the MM-Fi benchmark, plus the deployment mechanism that resolves real-world generalization:
- **Efficiency frontier** — a **75 K-param model beats published SOTA** (74.3% vs MultiFormer 72.25% torso-PCK@20); every config from `micro` up is Pareto-dominant (smaller *and* more accurate than prior work). Shipped a deployable **int4 edge model (~20 KB, verified 74.08%, 0.135 ms single-thread CPU)** — published at [`ruvnet/wifi-densepose-mmfi-pose/edge`](https://huggingface.co/ruvnet/wifi-densepose-mmfi-pose). See [`docs/benchmarks/wifi-pose-efficiency-frontier.md`](docs/benchmarks/wifi-pose-efficiency-frontier.md).
- **Generalization solved by few-shot calibration** — zero-shot cross-subject (~64%) and cross-environment (~10%) are *not* closeable by algorithms (CORAL, DANN, instance-norm, contrastive foundation-pretraining all tested, all failed) or by more training subjects (saturates ~64%). But **~100–200 labeled in-room samples recover SOTA-level pose**: cross-subject 64→76%, **cross-environment 10→73% (60% from just 5 samples)** — deployable as a **~11 KB per-room LoRA adapter** on a frozen shared base. Full empirical chain in ADR-150 §3.2–3.6.
- **Calibration service** — Python reference impl (`aether-arena/calibration/`: `calibrate.py` fits an adapter from a labeled clip, `infer.py` runs base + adapter; verified 3.09%→74.29% on an unseen MM-Fi room) **integrated into the Rust product pose engine** (`cog-pose-estimation`): `InferenceEngine::with_adapter()` + `cog-pose-estimation run --config <cfg> --adapter <room.safetensors>` apply architecture-agnostic LoRA deltas on the pose head at inference. 6/6 crate tests pass (incl. an adapter-changes-output integration test).
- **ADR-147 — OccWorld world model integration** (`wifi-densepose-worldmodel` v0.3.0 published to crates.io). 15-frame trajectory prediction at 209 ms / 3.37 GB VRAM on RTX 5080. Phase 3 domain adapter `scripts/ruview_occ_dataset.py` (`RuViewOccDataset`) converts WorldGraph snapshots to OccWorld tensors with indoor class remapping + zero ego-poses (validated). Phase 5 retraining pipeline `scripts/occworld_retrain.py` — VQVAE + transformer fine-tuning on RuView occupancy snapshots. See [ADR-147](docs/adr/ADR-147-nvidia-cosmos-world-foundation-model-integration.md) · [benchmark proof](docs/adr/ADR-147-benchmark-proof.md). - **ADR-147 — OccWorld world model integration** (`wifi-densepose-worldmodel` v0.3.0 published to crates.io). 15-frame trajectory prediction at 209 ms / 3.37 GB VRAM on RTX 5080. Phase 3 domain adapter `scripts/ruview_occ_dataset.py` (`RuViewOccDataset`) converts WorldGraph snapshots to OccWorld tensors with indoor class remapping + zero ego-poses (validated). Phase 5 retraining pipeline `scripts/occworld_retrain.py` — VQVAE + transformer fine-tuning on RuView occupancy snapshots. See [ADR-147](docs/adr/ADR-147-nvidia-cosmos-world-foundation-model-integration.md) · [benchmark proof](docs/adr/ADR-147-benchmark-proof.md).
### Added ### Added