FreDSNet: Joint Monocular Depth and Semantic Segmentation with Fast Fourier Convolutions
Open paper
Benchmark
Model
Rank
Results
depth-estimation-on-stanford2d3d-panoramic
FreDSNet
#1
RMSE: 0.2727
absolute relative error: 0.0952
semantic-segmentation-on-stanford2d3d-1
FreDSNet
#17
mIoU: 46.1%
mAcc: 63.1
Rank counts only results with a code link.