Open Access

表1

DTU数据集上的评估结果

分类 方法 平均距离/mm
精度 完整度 整体质量
传统几何方法 Furu[20] 0.613 0.941 0.777
Camp[21] 0.835 0.554 0.695
Gipuma[5] 0.283 0.873 0.578
Colmap[6] 0.400 0.664 0.532
有监督方法 Surfacenet[22] 0.450 1.040 0.745
MVSNet[1] 0.444 0.741 0.592
R-MVSNet[3] 0.383 0.452 0.417
Fast-MVSNet[23] 0.336 0.403 0.370
PatchmatchNet[10] 0.427 0.277 0.352
CVP-MVSNet[4] 0.296 0.406 0.351
无监督方法 Unsup-MVS[11] 0.881 1.073 0.977
MVS2[24] 0.760 0.515 0.637
M3VSNet[12] 0.636 0.531 0.583
JDACS[19] 0.571 0.515 0.543
Uns-PMVSNet 0.436 0.515 0.476

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.