Measurement-level LiDAR-inertial-visual odometry can be accurate, but persistently attaching dense photometric patches or appearance attributes to mapped LiDAR points adds storage, association, and update costs. F-LIVO instead relies on sparse geometric features. It is a filter-based framework that couples a visual-inertial subsystem and a LiDAR-inertial subsystem through feature-level point-to-plane constraints between visual points and LiDAR planes.

To preserve cross-modal coupling when visual triangulation is weak, F-LIVO recovers visual feature depth from the LiDAR map through local plane fitting and ray-plane intersection. In the LiDAR update, the covariance of each point-to-plane residual is adaptively scaled according to local planar reliability, reducing overconfident updates on poorly conditioned surfaces. Because the method avoids persistent dense photometric map attributes, it keeps a compact map representation.

Experiments on multiple public datasets show competitive, sequence-dependent accuracy. Across evaluated degradation cases, F-LIVO maintains continuous estimation with comparatively low accumulated end-to-end drift. Under the evaluated resource configuration, it achieves the lowest mean and peak CPU usage, the lowest peak resident set size, and the lowest RSS growth rate among the evaluated implementations, while R2LIVE retains a lower mean RSS. The source code will be made publicly available upon acceptance.