# Low-tech method: Drawing smooth tunnel centerlines with constant node spacing in JOSM using a physical circle overlay **URL:** \ **Category:** Uncategorized\ **Created:** [March 15, 2026, 6:29am UTC](https://forum.mapillary.com/t/low-tech-method-drawing-smooth-tunnel-centerlines-with-constant-node-spacing-in-josm-using-a-physical-circle-overlay/10263 "2026-03-15T06:29:45Z")\ **Posts on this page:** 1\ **Page:** 1
**Author:** ![osmplus\_org](https://sea1.discourse-cdn.com/flex015/user_avatar/forum.mapillary.com/osmplus_org/32/72465_2.png) [@osmplus\_org](https://forum.mapillary.com/u/osmplus_org)\ **Post date:** [March 15, 2026, 6:29am UTC](https://forum.mapillary.com/t/low-tech-method-drawing-smooth-tunnel-centerlines-with-constant-node-spacing-in-josm-using-a-physical-circle-overlay/10263/1 "2026-03-15T06:29:45Z")
**Low-tech method: Drawing smooth tunnel centerlines with constant node spacing in JOSM using a physical circle overlay** Hello everyone, I would like to share a very simple, low-tech technique I use to manually draw tunnel centerlines (or other strongly curved ways) in JOSM with **almost perfectly constant node spacing** — usually 20–35 meters — even on tight curves. This helps Mapillary interpolate the sequence very smoothly without visible bunching or stretching. The method requires **no plugins** , no scripts and almost no money — just a transparency sheet, a compass and JOSM’s built-in auto-centering. ### How it works – step by step 1. **Create a physical reference circle** 2. **Attach the foil to the screen** 3. **JOSM setup** 4. **Drawing the tunnel way** → Result: you get a very regular node spacing even on strong bends, while still being able to closely follow the visual tunnel path or the OSM way. ### Advantages - Geometrically exact spacing (real circle, not algorithm approximation) - Complete visual control – you decide where the centerline really belongs - Zero software dependencies → works with any future JOSM version - Very cheap (~3–8 € for foil + compass) - Easy to teach to other contributors ### Important note about GPX export for Mapillary When you export such a hand-drawn way as GPX, the file naturally contains only the coordinates of the nodes you placed — **no timestamps** are included. To make the GPX usable for Mapillary (which relies on time information for correct sequencing and speed estimation), you still need to calculate / interpolate reasonable timestamps afterwards. For one possible approach to fill in the missing times see this older thread: > [@No more GPS stress thanks to navigation using video frames](https://forum.mapillary.com/t/no-more-gps-stress-thanks-to-navigation-using-video-frames/10152/6): > > Here is an example of how we can improve—or even completely reconstruct—GPX recordings. Let’s take a closer look at the GPX dataset I recorded during a tour of Red Bull Hangar 7. Dataset 1 shows a GPX recording that is highly inaccurate due to poor GPS reception caused by the building’s glass roof. Systems such as Insta360 or Osmo360 tend to generate extremely dense GPX data, which later results in oversampling on Mapillary. Example file 1 was generated by an Osmo360. Dataset 2 shows a GPX fil… ### Limitations - Manual clicking → slower than automatic tools - Requires initial calibration (circle size + zoom level) ### Help - Any minor irregularities that appear on long straight parts can easily be smoothed out - afterwards using JOSM’s built-in Shift + B function (Align Nodes in Line). Has anyone tried similar low-tech tricks or even improved variations? I’m especially interested in: - How other people determine / calibrate the correct circle radius for different screens - Whether printing transparent radial patterns or protractor-like overlays would be even more convenient - Experiences with constant spacing vs. variable spacing for Mapillary interpolation quality Looking forward to your feedback and possible refinements! Best regards osmplius\_org