Initial depth estimation using EPIs and structure tensor

2017 ◽  
Author(s):  
Wisarut Chantara ◽  
Yo-Sung Ho
Keyword(s):  
Depth Estimation ◽  
Structure Tensor

Real-time Depth Estimation and View Interpolation using Quasar

2016 ◽  
Vol 2016 (19) ◽  
pp. 1-6 ◽  
Author(s):  
Bart Goossens ◽  
Simon Donné ◽  
Jan Aelterman ◽  
Jonas De Vylder ◽  
Dirk Van Haerenborgh ◽  
...  
Keyword(s):  
Real Time ◽  
Depth Estimation ◽  
View Interpolation

Method for Automatic Switching Screen of OST-HMD using Gaze Depth Estimation

2018 ◽  
Vol 7 (1) ◽  
pp. 31-36
Author(s):  
Youngho Lee ◽  
◽  
Choonsung Shin ◽  
Keyword(s):  
Depth Estimation ◽  
Automatic Switching

Unsupervised Monocular Depth Estimation for Autonomous Driving

2019 ◽  
pp. 128
Author(s):  
Chih-Shuan Huang ◽  
Wan-Nung Tsung ◽  
Wei-Jong Yang ◽  
Chin-Hsing Chen
Keyword(s):  
Depth Estimation ◽  
Autonomous Driving ◽  
Monocular Depth

Omni-Directional Depth Estimation System with Single-Double Mirrors Architecture

2010 ◽  
Vol 22 (6) ◽  
pp. 1004-1009 ◽  
Author(s):  
Wang Chen ◽  
Maojun Zhang ◽  
Yang Chong ◽  
Zhihui Xiong
Keyword(s):  
Depth Estimation ◽  
Estimation System

scholarly journals Improved Automated Detection of Subpixel-Scale Inundation—Revised Dynamic Surface Water Extent (DSWE) Partial Surface Water Tests

2019 ◽  
Vol 11 (4) ◽  
pp. 374 ◽  
Author(s):  
John Jones
Keyword(s):  
Surface Water ◽  
Mixture Models ◽  
Model Development ◽  
Depth Estimation ◽  
High Temporal Resolution ◽  
Commission Error ◽  
Dynamic Surface ◽  
In Situ Data ◽  
The Us ◽  
Spectral Mixture

In order to produce useful hydrologic and aquatic habitat data from the Landsat system, the U.S. Geological Survey has developed the “Dynamic Surface Water Extent” (DSWE) Landsat Science Product. DSWE will provide long-term, high-temporal resolution data on variations in inundation extent. The model used to generate DSWE is composed of five decision-rule based tests that do not require scene-based training. To allow its general application, required inputs are limited to the Landsat at-surface reflectance product and a digital elevation model. Unlike other Landsat-based water products, DSWE includes pixels that are only partially covered by water to increase inundation dynamics information content. Previously published DSWE model development included one wetland-focused test developed through visual inspection of field-collected Everglades spectra. A comparison of that test’s output against Everglades Depth Estimation Network (EDEN) in situ data confirmed the expectation that omission errors were a prime source of inaccuracy in vegetated environments. Further evaluation exposed a tendency toward commission error in coniferous forests. Improvements to the subpixel level “partial surface water” (PSW) component of DSWE was the focus of this research. Spectral mixture models were created from a variety of laboratory and image-derived endmembers. Based on the mixture modeling, a more “aggressive” PSW rule improved accuracy in herbaceous wetlands and reduced errors of commission elsewhere, while a second “conservative” test provides an alternative when commission errors must be minimized. Replication of the EDEN-based experiments using the revised PSW tests yielded a statistically significant increase in mean overall agreement (4%, p = 0.01, n = 50) and a statistically significant decrease (11%, p = 0.009, n = 50) in mean errors of omission. Because the developed spectral mixture models included image-derived vegetation endmembers and laboratory spectra for soil groups found across the US, simulations suggest where the revised DSWE PSW tests perform as they do in the Everglades and where they may prove problematic. Visual comparison of DSWE outputs with an unusual variety of coincidently collected images for locations spread throughout the US support conclusions drawn from Everglades quantitative analyses and highlight DSWE PSW component strengths and weaknesses.

scholarly journals Unsupervised Learning-Based Depth Estimation-Aided Visual SLAM Approach

2019 ◽  
Vol 39 (2) ◽  
pp. 543-570 ◽  
Author(s):  
Mingyang Geng ◽  
Suning Shang ◽  
Bo Ding ◽  
Huaimin Wang ◽  
Pengfei Zhang
Keyword(s):  
Unsupervised Learning ◽  
Depth Estimation ◽  
Visual Slam
Sign in / Sign up

Export Citation Format

Share Document