OpenLiDARViewer v0.5.5

A navigation, interface, reporting, and validation release. v0.5.5 adds a Pan hand tool, refines viewport navigation and point rendering, makes the side panels collapsible, simplifies the PDF report set, corrects scan-health reporting for deliberately sampled datasets, and adds reproducible performance diagnostics.

OpenLiDARViewer stays browser-native and local-first. Local files never leave the device, and no account is required.

Pan hand tool

A fourth navigation mode, Pan, lets you grab the scene and slide it across the viewport. Activate it from the navigation control, or:

• 4 selects Pan;
• G toggles Pan on and off;
• middle-mouse drag pans temporarily from any other mode.

In Pan mode a primary mouse or pen drag moves the scene while the wheel keeps zooming. One-finger touch dragging works too. Camera orientation and view scale stay fixed during the drag. Pan mode is preserved in saved sessions and share links.

Viewport navigation and rendering

Wheel and trackpad zoom is now frame-rate independent: the same gesture reaches the same zoom on a 60, 120, or 144 Hz display. Zoom also centres on the pointer, so the point under the cursor stays put. The previous behaviour stays available with ?wheelDolly=legacy.

While you move the view, the renderer lowers the device-pixel ratio a little and restores it once you stop, which keeps interaction smooth on dense scenes without a visible drop in sharpness. Turn it off with ?adaptiveDpr=off.

A Gaussian point-appearance mode joins the existing point styles. It softens ordinary point samples; it is not a trained 3D Gaussian Splat scene.

Collapsible side panels

The left panel column collapses with a one-tap handle. On the right, the Inspector and (when a COPC dataset streams) the streaming card each collapse on their own handle, so you can hide either without the other. Every handle remembers its state per browser and stays hidden until a scan is open; on small screens the panels move into the bottom sheet instead. A wheel over any panel scrolls only that panel and never moves the camera.

Simpler PDF report set

The overlapping report catalogue is now two documents:

• Survey Summary: a compact handover with the inspection summary, dataset information, concise provenance, measurements, and any supplied technical notes.
• Technical Report: the full record, adding detailed provenance, file-declared source metadata, annotations, and visuals.

Older report-template identifiers map to the nearest current template, so existing sessions and integrations still open.

The Scan Acceptance template is gone. Its metadata-presence rows did not amount to an acceptance test; acceptance reporting should return only when it is backed by explicit, data-derived checks and user-defined criteria.

More accurate scan-health reporting

The Health Check now separates three cases: a complete decode, a deliberate display-sample cap, and a declared-versus-decoded count mismatch. A large LAS or LAZ file loaded with an intentional sampling stride reads as sampled instead of being flagged as having lost points.

The decoded count and the applied stride now cross the parsing-worker boundary intact, so the interface reports the same loading state the decoder saw.

Smaller report fixes: classification coverage shows in the Classification row, repeated analysis caveats are merged, and an empty cloud reports a verdict instead of an unrelated point count.

Reproducible performance diagnostics

The optional debug overlay now records frame-time percentiles (p50/p95/p99), counts of frames over common frame-time thresholds, the longest observed main-thread task where the browser supports it, the effective device-pixel ratio, and rendering and streaming counters. The values copy out as JSON for before-and-after comparisons.

A deterministic scheduler baseline is included for regression testing. These are measurement tools; the release does not claim a general rendering or streaming speedup without device-specific evidence.

Compatibility and scope

Orbit, Walk, and Fly navigation are unchanged, as are the existing point-cloud rendering modes and the local-first, static-host deployment model. Legacy report-template identifiers fall back to supported templates.

Known limitations

• Pan moves the camera; it does not move or edit point-cloud coordinates.
• Performance metrics depend on the browser, device, viewport, and dataset, and are not universal benchmarks.
• Reports summarize the available data and implemented analyses. They are not survey certification or independent validation.
• A deliberately sampled load stays a display sample, not a full in-memory copy of the source cloud.
• The Gaussian point-appearance mode shapes ordinary point samples. Loading a trained 3D Gaussian Splat scene is not in this release.

Verify this release

npm ci
npm run test:release

To check the main v0.5.5 changes directly:

npx vitest run \
  tests/panMath.test.ts \
  tests/wheelDollyMath.test.ts \
  tests/adaptiveDpr.test.ts \
  tests/refinementPhase.test.ts \
  tests/splatShader.test.ts \
  tests/frameTelemetry.test.ts \
  tests/metricsJson.test.ts \
  tests/reportTemplateGoldens.test.ts \
  tests/healthCheck.test.ts \
  tests/scanReport.test.ts

See CHANGELOG.md for the implementation-level history.

Deploy

Static files. Host on GitHub Pages, Netlify, a static CDN, or any conventional web host.