Filters








6 Hits in 1.4 sec

ResDepth: Learned Residual Stereo Reconstruction [article]

Corinne Stucker, Konrad Schindler
2021 arXiv   pre-print
The strategy to only learn the residual greatly simplifies the learning problem.  ...  We propose an embarrassingly simple but very effective scheme for high-quality dense stereo reconstruction: (i) generate an approximate reconstruction with your favourite stereo matcher; (ii) rewarp the  ...  ResDepth-stereo further improves the reconstruction.  ... 
arXiv:2001.08026v3 fatcat:w5sibck3ive7bnefqcp7iicwze

ResDepth: A Deep Residual Prior For 3D Reconstruction From High-resolution Satellite Images [article]

Corinne Stucker, Konrad Schindler
2021 arXiv   pre-print
Modern optical satellite sensors enable high-resolution stereo reconstruction from space.  ...  To that end, we introduce ResDepth, a convolutional neural network that learns such an expressive geometric prior from example data.  ...  Here, we show that a residual learning strategy greatly simplifies the learning problem.  ... 
arXiv:2106.08107v2 fatcat:cvvsxqyn4ngcbbd4gd7qb7fhua

IMPROVING DISPARITY ESTIMATION BASED ON RESIDUAL COST VOLUME AND RECONSTRUCTION ERROR VOLUME

J. Kang, L. Chen, F. Deng, C. Heipke
2020 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
The main contribution of this paper is to combine the residual cost volume and the reconstruction error volume to guide training of the refinement network.  ...  We evaluate our method on several challenging stereo datasets.  ...  Most recently, ResDepth, a deep network (Stucker and Schindler, 2020) was proposed to improve the depth map for high-quality dense stereo reconstruction.  ... 
doi:10.5194/isprs-archives-xliii-b2-2020-135-2020 fatcat:xxjlnaj7orexhjfgggrcs23fkq

MetaMQAP: A meta-server for the quality assessment of protein models

Marcin Pawlowski, Michal J Gajda, Ryszard Matlak, Janusz M Bujnicki
2008 BMC Bioinformatics  
As a reference, we calculated the value of correlation between the local deviations and trivial features that can be calculated for each residue directly from the models, i.e. solvent accessibility, depth  ...  eight MQAPs: VERIFY3D, PROSA, BALA, ANOLEA, PROVE, TUNE, REFINER, PROQRES on 8251 models from the CASP-5 and CASP-6 experiments, by calculating the Spearman's rank correlation coefficients between per-residue  ...  This procedure reconstructed a heavy-atom representation for all residues except the omitted terminal residues and optimized the bond lengths, angles and packing.  ... 
doi:10.1186/1471-2105-9-403 pmid:18823532 pmcid:PMC2573893 fatcat:olfvxgorjjaxvdzya37i6dgqgu

CVPRW 2020 TOC

2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
A Training Method for Image Compression Networks to Improve Perceptual Quality of Reconstructions 585 Jooyoung Lee (Broadcasting and Media Research Laboratory, Electronics and Telecommunications Research  ...  : Learned Residual Stereo Reconstruction 707 Corinne Stucker (Photogrammetry and Remote Sensing, ETH Zurich, Switzerland) and Konrad Schindler (Photogrammetry and Remote Sensing, ETH Zurich, Switzerland  ...  Attribute-Conditioned Synthesis 645 Zac Yu (University of Pittsburgh) and Adriana Kovashka (University of Pittsburgh) xvii EarthVision: EarthVision: Large Scale Computer Vision for Remote Sensing Imagery ResDepth  ... 
doi:10.1109/cvprw50498.2020.00004 fatcat:6qao4eypyvg3xiiigih3kicniq

Improving disparity estimation based on residual cost volume and reconstruction error volume [article]

Junhua Kang, Lin Chen, Fei Deng, Christian Heipke, University, My, N. Paparoditis, C. Mallet, F. Lafarge, Fabio Remondino, Isabella Toschi, Takashi Fuse
2021
The main contribution of this paper is to combine the residual cost volume and the reconstruction error volume to guide training of the refinement network.  ...  We evaluate our method on several challenging stereo datasets.  ...  Most recently, ResDepth, a deep network (Stucker and Schindler, 2020) was proposed to improve the depth map for high-quality dense stereo reconstruction.  ... 
doi:10.15488/10823 fatcat:6izsrmmdybhb7mxd45azkrw5tq