Sparse and Dense Data with CNNs: Depth Completion and Semantic Segmentation

Convolutional neural networks are designed for dense data, but vision data is often sparse (stereo depth, point clouds, pen stroke, etc.). We present a method to handle sparse depth data with optional dense RGB, and accomplish depth completion and semantic segmentation changing only the last layer. Our proposal efficiently learns sparse features without the need of an additional validity mask. We show how to ensure network robustness to varying input sparsities. Our method even works with densities as low as 0.8% (8 layer lidar), and outperforms all published state-of-the-art on the Kitti depth completion benchmark.

PDF Abstract

Results from Other Papers

Task Dataset Model Metric Name Metric Value Rank Source Paper Compare
Depth Completion KITTI Depth Completion Spade-sD RMSE 1035 # 10
MAE 248 # 7
Runtime [ms] 40 # 9
Depth Completion KITTI Depth Completion Spade-RGBsD RMSE 918 # 8
MAE 235 # 6
Runtime [ms] 70 # 10


No methods listed for this paper. Add relevant methods here