2019 · 30 citations · 5 references
ComputingComputer Science3D ModelingComputer VisionMachine Vision3D ReconstructionDeep LearningDepth EstimationLoss FunctionScene ReconstructionVisual SlamConvolutional Neural NetworksDense 3D ReconstructionDense ReconstructionDepth Map EstimationDepth Map ReconstructionPerceptual LossSparse 3D ReconstructionSparse Reconstruction
In our recent papers, we proposed a new family of residual convolutional neural networks trained for semi-dense and sparse depth reconstruction without use of RGB channel. The proposed models can be used in low-resolution depth sensors or SLAM methods estimating partial depth with certain distributions. We proposed using perceptual loss for training depth reconstruction in order to better preserve edge structure and reduce over-smoothness of models trained on MSE loss alone. This paper contains reproducibility companion guide on training, running and evaluating suggested methods, while also presenting links on further studies in view of reviewers comments and related problems of depth reconstruction.
5
On Learning Conditional Random Fields for Stereo
Christopher Pal, Jerod Weinman, Lam C. Tran et al. · International Journal of Computer Vision · 2010 · 30 citations
Fast Semi-dense Depth Map Estimation
Ilya Makarov, Alisa Korinevskaya, Vladimir Aliev · 2018 · 26 citations
[POSTER] Depth Map Interpolation Using Perceptual Loss
Ilya Makarov, Vladimir Aliev, Olga Gerasimova et al. · 2017 · 25 citations