@Nik__V__
Meet MapAnything ā a transformer that directly regresses factored metric 3D scene geometry (from images, calibration, poses, or depth) in an end-to-end way. No pipelines, no extra stages. Just 3D geometry & cameras, straight from any type of input, delivering new state-of-the-art results š One universal model enables SoTA for: š„ Mono Depth Estimation š„ Multi-View SfM š„ Multi-View Stereo š„ Depth Completion š„ Registration ⦠and many more possibilities! ā plus everything is metric šÆ We release code for data processing, training, benchmarking & ablations ā everything Apache 2.0! Details & Links š