Hi,
I want to reuse an existing .sfm file (with known intrinsic and extrinsic calibration from a camera rig) to run a photogrammetry reconstruction on just two images, without re-running the Structure-from-Motion (SfM) step.
Specifically, I want to apply the locked poses and intrinsics to skip camera pose estimation and go directly to downstream steps like FeatureMatching, DepthMap, and Texturing.
Is there a recommended way to inject the locked camera poses and proceed directly to mesh or depth map generation (for example in Meshroom or another photogrammetry tool)?
Thanks!