Manydepth2: Motion-Aware Self-Supervised Multi-Frame Monocular Depth Estimation in Dynamic Scenes

Benchmark Model Rank Results
camera-pose-estimation-on-kitti-odometryManydepth2#1Average Translational Error et[%]: 7.15
monocular-depth-estimation-on-cityscapesManydepth2#2RMSE: 5.827RMSE log: 0.154Absolute relative error (AbsRel): 0.097
monocular-depth-estimation-on-kitti-eigenManydepth2#44absolute relative error: 0.091RMSE: 4.232Sq Rel: 0.170
monocular-depth-estimation-on-kitti-eigen-1Manydepth2(M+640x192)#6absolute relative error: 0.091RMSE: 4.232RMSE log: 0.170
monocular-depth-estimation-on-kitti-eigen-1Manydepth2-NF(M+640x192)#10absolute relative error: 0.094RMSE: 4.246RMSE log: 0.170