Turns detector output into queryable relations plus a symbolic rule engine, with plain-language explanations, on a zero-dependency core that runs on Jetson-class devices as-is.
Closed source, inquireEdge-scene-graphs sits between any detector and an alerting system: geometric spatial and temporal predicate extractors (near, overlaps, in a zone, loitering, crossed a line, and more), an interval-compressed scene graph, and a declarative rule engine (joins, sequences, absence, cooldowns) that produces grounded, explainable alerts plus an ad-hoc relation query language.
The core is Python-standard-library only, no numpy, no torch, so it runs unmodified on Jetson-class hardware next to a TensorRT detector.
It also ships a full learned-relation-head path: a public scene-graph benchmark data converter, training tooling, a trained torch MLP checkpoint, and a Recall-at-K evaluation harness.
A scripted 60-second, 10-frames-per-second scenario processed 600 frames into 12,454 triples and 54 intervals, firing 10 alerts across 3 rule types, in 0.1357 seconds (about 4,420 frames per second processed).
Mean 1.981 ms per frame, p95 2.233 ms, capacity about 504.7 frames per second.
| Objects/frame | Mean ms | p50 ms | p95 ms | fps capacity |
|---|---|---|---|---|
| 6 | 2.282 | 2.443 | 2.663 | 438.3 |
| 12 | 9.926 | 9.905 | 13.435 | 100.7 |
| 20 | 25.35 | 25.273 | 31.552 | 39.4 |
| R@20 | mR@20 | R@50 | mR@50 | |
|---|---|---|---|---|
| Trained head | 0.500 | 0.159 | 0.596 | 0.200 |
| Class-pair frequency baseline | 0.220 | 0.061 | 0.390 | 0.118 |
| Geometric baseline (vocab mismatch) | 0.007 | 0.014 | - | - |
Trained on an RTX 5070 Ti, 30 epochs, about 2 minutes. The same head, exported to ONNX and built into a TensorRT 10.3 FP16 engine, runs on an Orin Nano Super at 0.117 ms mean and 0.122 ms p95 per batch-64 inference (about 9,404 queries per second), a 334 KB engine file, 0.078 MiB of execution-context device memory.
Today an outsider cannot reproduce any of this hands-on: there is no public dataset, no demo Space, and the source is closed. The numbers above, including the scene-graph benchmark evaluation and the Jetson latency table, are reported from the project's own results files, not independently verifiable from outside yet.