3Dnovator Model the world around you

Reconstruct and measure 3D scenes from photos

3Dnovator is a young company specialized in researching and developing Structure-from-Motion (SfM) solutions. Our main software solution, an advanced image-based 3D modeling library, aims at creating professional quality 3D content from still images in an easy and automatic way, without the necessity of any specialized knowledge or hardware. Based on the latest multi-view 3D reconstruction technology, it operates with arbitrary images and is efficient in both controlled and uncontrolled conditions. Photos can be taken from any positions, providing that an object to be reconstructed is visible in at least three photos. Both image alignment and 3D model reconstruction are fully automated. The entire range of the photogrammetry operations, like distance and volume measurements, are now at only one mouse click away.

Main features

Sparse reconstruction:

  • Fast and accurate image matching for both small and large photo-sets
  • Support video input: automatic frames extraction and features tracking
  • Support for man-made structures through line detection and matching
  • False image pairs detection system (problem often encountered during man-made scenes reconstruction)
  • Fix or automatically estimated camera intrinsics and extrinsics
  • Heterogeneous cameras system support: central and non-central (multi-camera rig)
  • Fast and robust reconstruction pipeline with support for automatic loop-closure
  • Auto-rotates reconstruction (detects upwards direction)
  • Support GPS and GCP (Ground Control Points) info, if available, for geo-registration and reconstruction optimization
  • Large photo-set support through clustering
  • Perform real world measurements like distance and volume computations

Dense reconstruction:

  • Fast dense point cloud reconstruction
  • Automatic mesh reconstruction and texturing
  • Mesh refinement