OpenMVG

OpenMVG is a C++ toolkit for building image-based 3D reconstruction workflows, aimed chiefly at developers and researchers. Its open-source libraries cover geometry and structure from motion, but producing dense point clouds or textured meshes means taking the reconstructed scene into another framework.

Overview

OpenMVG brings together libraries, command-line binaries, and pipelines for multiple-view geometry and structure from motion. Its scope runs from image manipulation and feature processing through camera models and robust estimation to reconstruction and scene export. That breadth is useful when you need control over how a workflow is assembled; it is less suitable if you want a ready-made scanning app with a consumer capture experience.

The project is written in C++ and targets reproducible research through readable implementations of classic and state-of-the-art algorithms. Test-driven development and samples support learning and modification, while tutorials cover data structures, library use, and processing an image dataset. Those are meaningful advantages for technical users, though the emphasis on libraries and pipelines assumes more development involvement than a turnkey tool.

Key features

From images to a reconstructed scene

OpenMVG’s libraries cover image manipulation, feature description and matching, feature tracking, camera models, multiple-view geometry, robust estimation, and structure-from-motion algorithms. Its binaries handle scene initialization, feature detection and matching, reconstruction, and export. Together, these pieces provide a substantial base for custom image-to-scene workflows rather than locking users into one reconstruction step.

Export for dense reconstruction

OpenMVG can export reconstructed scenes to other multiple-view-stereo frameworks, which can then compute dense point clouds or textured meshes. That separation is a practical boundary: OpenMVG supplies the geometry and reconstruction pipeline, but users seeking those dense outputs must bring another framework into their process.

Research and adaptation

Readable implementations, samples, and test-driven development make the project a reasonable choice for studying or adapting reconstruction methods. Tutorials address both library use and dataset processing, and the project points users to its wiki and a contact email for usage resources and support. The research orientation is a strength for people who want to understand or change the underlying methods, not a substitute for a packaged end-user interface.

Pricing

OpenMVG is free and open source under the Mozilla Public License 2.0. Its OpenMVG plan costs nothing and includes an open-source library under MPL-2.0. There is no paid tier to unlock the core toolkit, making it suitable for users who can work directly with the software rather than pay for a managed scanning service.

Platforms

OpenMVG is categorized for Android, iOS, Linux, macOS, self-hosted use, and Windows, with desktop deployment. Its C++ library and pipeline orientation makes it a software development choice rather than a mobile-first capture app.

Who it's for

Choose OpenMVG if you are a developer or researcher who wants to build, inspect, or modify image-based reconstruction workflows, or needs a free foundation for structure-from-motion work. It is a weaker fit for people who need dense point clouds or textured meshes directly from a single app, or who prefer a finished capture-and-export experience without assembling a pipeline.

Pros and cons

• Pros: A broad set of geometry, feature-processing, and structure-from-motion components supports custom workflows from image inputs through scene export.
• Pros: Free access under MPL-2.0 removes a software subscription cost for developers and researchers.
• Pros: Readable implementations, samples, and tutorials give technical users material for learning and adapting the algorithms.
• Cons: Dense point clouds and textured meshes require another multiple-view-stereo framework, adding a dependency for users who need those outputs.
• Cons: Its library-and-pipeline focus is a poor match for users seeking a ready-made scanning interface rather than a toolkit to work with.

Alternatives

Photogrammetry Software and AI 3D Scanning Software are useful category pages for comparing other approaches.

• COLMAP is another free option for users comparing image-based reconstruction software.
• MicMac is a free, open-source photogrammetric suite for readers seeking a broader photogrammetry-oriented alternative.
• WebODM is free to install and use locally and offline, making it a better alternative for users who want a locally run mapping workflow.
• MagiScan offers a freemium option for people looking for a scanning-oriented alternative across mobile and desktop platforms.
• PIX4Dcatch is a freemium capture alternative with LiDAR 3D capture, GPS georeferencing, point measurement, and GCP creation in its free Discovery license.
• KIRI Engine offers a free Basic plan with up to 150 photos, a 2 GB per-scan limit, unlimited scans, and free exports, suiting users who want defined scan limits and exports in a freemium tool.
• MetaScan has a free tier capped at five scans in the library and one HD reconstruction per month, an option for users who can work within those limits.
• Polycam offers a free tier with limited space, object, and AI captures, plus GLTF exports and public link sharing, for users who want a capture-focused alternative.

Verdict

OpenMVG is a strong choice for developers and researchers who want a free, adaptable foundation for image-based 3D reconstruction and value access to the underlying algorithms. Look elsewhere if your priority is a finished scanning app or dense point clouds and textured meshes without adding a separate multiple-view-stereo framework.