UAV3DCrop: Benchmarking 3D Reconstruction in Repeated Multi-Angle UAV Crop Surveys
This work introduces UAV3DCrop, a public benchmark of repeated multi-angle unmanned aerial vehicle (UAV) crop surveys, and evaluates seven scene-optimized methods -- Neural Radiance Field and 3D Gaussian Splatting variants -- on held-out views, photogrammetry-referenced depth, and canopy-height recovery.