# Cloud processing and a desktop editor, built as one. > Everything CloudODM and ODMDisplay do: photogrammetry and LiDAR classification in the cloud, and a desktop editor to measure, correct classes and report stockpiles on the same survey. Page: https://cloudodm.stellaris.cl/features ## The price before the run, and exactly that price A fixed amount per photograph or per million points, from your own files, shown before anything starts and charged to the cent when it completes. No subscription to process, no credits that expire, no meter. A failed run costs nothing, and two Basic runs of up to 300 photographs are free each month. 400 photographs, map and 3D: US$8.00 on Basic. ## Cloud processing and a desktop editor that share one survey Run the heavy work in the cloud, open the full-resolution result on your own machine, correct it, and send it back. Datasets stay in step between the two with checks on every file, and measurements made on either side appear on the other for the whole team. ## Classification in three tiers, with the numbers published Instant geometric classification on your own computer, a standard cloud classifier for every run, and a deep-learning tier for scenes that have to be right. Each was scored point by point against a professional hand-checked reference, and the scores are on the page. Deep-learning tier: 95.6% whole-scene accuracy, buildings 83% (standard: 19%). ## Terrain models that leave out large buildings and keep steep ground The usual ground filters fail in opposite directions: one keeps stadiums and warehouse roofs as terrain, the other shaves mountainsides. The default combines them and falls back safely on steep sites — measured on a flat campus, a mountain park and a hillside subdivision. Autzen terrain model error: 0.41 m → 0.08 m. ## Files checked before you pay for anything Your images are read in the browser before any space is bought or byte uploaded: stray files that would break a run, photographs the engine cannot decode and colliding file names are named, and the run is priced from the real files. ## An assistant can drive it The desktop viewer runs an MCP service: an AI assistant can open a cloud, move around it, classify, measure and export, and answer questions about the site in plain language. ## Open formats, and your coordinate system kept GeoTIFF, cloud-optimised GeoTIFF, LAS, LAZ, COPC, GLB, DXF, GeoJSON and LandXML — nothing locked in. Surveys in state-plane grids and feet are kept in their own system and still placed on the map. ## Everything it does ### Photogrammetry in the cloud - Orthophoto, surface model, terrain model, point cloud and textured 3D model from drone photographs - Contour lines at an interval fitted to the site's relief - Ground classification and AI scene classification of the reconstructed cloud - Adjustable detail: photo alignment, point cloud, mesh, orthophoto and elevation resolutions - Orthophoto-only fast mode - Streamable outputs: COPC point cloud, cloud-optimised GeoTIFFs, 3D model for the web - Processing report with alignment quality ### LiDAR in the cloud - Classify-only jobs for LAS, LAZ, E57, PLY, PCD and BPF — no photographs needed - Ground filters: Automatic (SMRF checked by CSF), SMRF or CSF, with adjustable parameters - Standard classifier (ground, vegetation, buildings) and a deep-learning tier (adds power lines, poles, fences and vehicles) - Terrain and surface models, contours and a top-down orthophoto from the cloud's own colour - Metres and feet, and any coordinate system, handled automatically - A client's own classification kept exactly as delivered ### Price, payment and storage - Fixed price per photograph or per million points, shown before the run, charged to the cent - Failed runs free; two completed runs free each month; no cancel fee until a run is processing - Estimate of time, price and result size while you choose options - Prepaid balance or card; invoices - Storage bought by size and term; each run's outputs hosted for the days you choose (from five), priced in its quote - An email before outputs are deleted, as many days ahead as you set; more days bought in one click - Datasets deleted from the website or the desktop app, the local copy to the Recycle Bin - Every settlement checked by an independent second calculation ### Results in the browser - One viewer for point cloud, 3D model, orthophoto, surface and terrain models - Layers on a street map with opacity, including state-plane and feet surveys - Height under the pointer; contours with DXF and GeoJSON export - Length measurement with slope length, ascent and descent - Measurements shared with the desktop viewer - A check that every requested output was delivered, and every file to download ### Point management in the desktop viewer - Opens LAS, LAZ, COPC (streamed), E57, PLY, PCD, BPF and GeoTIFF; full-resolution clouds on your own machine - Filters by class, return, intensity, scan angle, GPS time and source - Local classification: ground (SMRF, CSF, morphological), height above ground, geometric scene classes, classes from DXF shapes - Paint and correct classes with a brush or polygon, with undo and redo - Noise and dust removal, spike removal, thinning - Datasets kept in step with the cloud: hash checks, versions, renames, edits sent back - Run details: dataset, options, time by stage, outputs and their hosting, price, bill and invoice PDFs - Save filtered or edited clouds as LAS/LAZ in a chosen coordinate system ### Viewing and measuring on the desktop - Orbit, Earth and Fly navigation; map view with drag and zoom at the pointer - Colour by RGB, elevation, intensity, class, return, GPS time, angle, NIR, source, density or change - Orthophoto, surface and terrain models on OpenStreetMap or Google, with opacity and contours - Surface height, terrain height and height above ground under the pointer - Distance, height, area, volume (cut and fill), angle, azimuth, circle and profile, with snapping - Stockpiles, cross-sections, breaklines, surfaces (TIN, Poisson) with slope, hillshade and curvature - Change between two epochs; planes and vectors traced from the cloud - Exports: DEM/DTM rasters, contours, TIN, LandXML, DXF, CSV, cross-sections, density and a report - Metres, feet and US survey feet; reprojection between UTM, geographic and web coordinates ### Teams and integration - Organisations with owners and members, invitations and roles - API tokens for scripts and integrations - An MCP service so an AI assistant can drive the desktop viewer - English and Spanish throughout ## More about CloudODM - [CloudODM: photogrammetry and LiDAR in the cloud](https://cloudodm.stellaris.cl/) - [Cloud photogrammetry](https://cloudodm.stellaris.cl/fotogrametria-en-la-nube) - [LiDAR classification](https://cloudodm.stellaris.cl/clasificacion-lidar) - [Point cloud viewer](https://cloudodm.stellaris.cl/visor-nube-de-puntos) - [Pricing](https://cloudodm.stellaris.cl/pricing) - [Measured accuracy](https://cloudodm.stellaris.cl/accuracy) - [Demonstration: a finished survey](https://cloudodm.stellaris.cl/demo) - [How it works](https://cloudodm.stellaris.cl/how) - [Tutorials](https://cloudodm.stellaris.cl/learn) - [ODMDisplay, the desktop application](https://cloudodm.stellaris.cl/viewer) - [Contact](https://cloudodm.stellaris.cl/contact) - [Terms and privacy](https://cloudodm.stellaris.cl/terms)