Appearance
Core concepts
Four ideas explain almost everything about how Kapnetix Post behaves. Read these once and the rest of the product follows.
Confidence is fixability
A confidence score predicts how cleanly Kapnetix can fix a take or a region — not merely that something is wrong. This is the single most important thing to internalise: confidence is never an error meter.
You see the same confidence map at two moments, and they are deliberately symmetric:
- Before (predicted). On the Triage screen, confidence forecasts how much review effort to expect, so you can plan the work before spending a token.
- After (measured). On the Review screen, confidence grades the result Kapnetix actually produced, so you can verify it.
If a result can't be measured, it's shown as Cleaned · unverified in the review queue rather than given a fabricated score. The banner on the After screen says which of five things the measurement was — Verified, Assessed with exclusions, Limited assessment, Not repairable or Not assessed — with the measured drivers underneath it. Only Verified is a clean bill of health.
Raw is immutable · Clean vN is non-destructive
Your raw capture is never modified. Every cleanup run stacks as a new named version — Clean v1, Clean v2, and so on — and the version history is never collapsed. You can always compare a clean version against the raw, or against an earlier clean, and you can revert individual regions.
A run is a versioned batch of takes cleaned together; a single take's history shows the run versions that produced it.
The four-state taxonomy
Every take and every region carries exactly one of four states. Before categorises the raw capture and signals the review effort to expect; After grades the result with the same colour and rank.
| State | Before label | Anticipated action | After label | Colour |
|---|---|---|---|---|
| Clean | Clean | No review needed | Cleaned ✦✦✦ | ● green |
| High confidence | High confidence | Minimal review | Cleaned ✦✦ | ● cyan |
| Low confidence | Low confidence | Review recommended | Cleaned ✦ | ● amber |
| Manual cleanup | Recommended manual cleanup | AI assists · you finish | Cleaned · finish manually | ● red |
What the states mean:
- Clean (green) — arrived clean, nothing to fix, kept as raw.
- High confidence (cyan) — auto-cleaned, high confidence in the result.
- Low confidence (amber) — cleaned, but with borderline regions worth a glance.
- Manual cleanup (red) — couldn't be auto-cleaned; the AI assists and you finish the pass.
Only Clean takes stay raw; everything else is scheduled for cleanup.
Confidence heat vs. status chips
The status chips above are categorical — one of four states. Separately, the brand pink → cyan gradient is the continuous per-frame confidence scale on the timeline and 3D viewer: cyan = high confidence (clean), pink = low (needs eyes), on a 0–100 scale where 0 means needs eyes and 100 means clean. Chips stay categorical; heat stays continuous.
Errors and the model that fixes each
Kapnetix classifies what it finds into five kinds, each repaired by a specific model:
| Error | Label on screen | What it is | Model |
|---|---|---|---|
| Gap | Gap / occlusion | Missing markers / occlusion | Solve · Gapfill |
| Merge | Marker merge | Two markers collapsed onto one | Solve · Gapfill |
| Swap | Marker swap | Marker labels swapped | Solve · Relabel |
| Jitter | Jitter | High-frequency noise | Solve · Denoise |
| Spike | Spike | A short, sharp impulsive peak | Solve · Denoise |
A class this build of the app cannot name shows as Other and claims no model, rather than being guessed at.
The order the work runs in
The principle does not change: complete the skeleton first, correct identities next, and treat noise last, once positions are right. A jitter cut runs after the fills of its own pass for the same reason — a cut has to rejoin the trajectory on both sides, and beside a hole there is nothing to rejoin to.
The exact order is worked out per take, from what the diagnosis found and what each step needs to be true first. That order is not hidden: it is the Cleanup Plan, and you can read it before you launch. See The Cleanup Plan.
In Auto mode Kapnetix picks the model per issue and plans the order for you. In Assisted mode you scope the work — frame range, error types, markers, solve constraints, jitter sensitivity — and the plan is built inside that scope. The governing rule throughout is the AI only touches what you scope.
What these ideas do not claim
- Confidence is not a promise. It is a prediction before the run and a measurement after it, and the two can disagree — that is the point of showing both.
- A Clean take is not a cleaned take. It arrived with nothing flagged and is kept as raw.
- A model name on a region card is not a guarantee that the fill shipped. A card that reads Nothing filled carries no model and no score, because nothing was fixed.
What to do next
Read how Post cleans a take for what happens between Before and After, or the quick start to walk it on a real file.
Ask us
Something looks wrong on a take? Write to support@kapnetix.ai and send the take id, or the take's name and the project it sits in.