en
0.25
0.5
0.75
1.25
1.5
1.75
2
Efficient Regression for Computational Photography: from Color Management to Omnidirectional Superresolution
Published on Jan 23, 20124096 Views
Many computational photography applications can be framed as low-dimensional regression problems that require fast evaluation of test samples for rendering. In such cases, storing samples on a grid or
Related categories
Chapter list
Efficient Regression for Computational Imaging: from Color Management to Omnidirectional Superresolution00:00
Regression - 100:04
Regression - 200:36
Regression - 300:56
Linear Regression: fast, not good enough01:34
Problem: Device Dependent Colors Depend on Device01:51
Color Management - 102:29
Color Management - 202:38
Color Management - 303:23
Gamut mapping: linear transforms not adequate03:40
Creating Custom Color Enhancements04:45
Example05:15
Color management: speed by LUT - 105:32
Color management: speed by LUT - 205:54
Color management: speed by LUT - 305:59
Color management: speed by LUT - 406:03
Color management: speed by LUT - 506:13
Color management: speed by LUT - 607:04
Color management: speed by LUT - 707:25
Linear Interpolation is linear in the outputs - 107:41
Linear Interpolation is linear in the outputs - 209:09
Linear Interpolation is linear in the outputs - 309:24
Lattice Regression - 110:05
Lattice Regression - 211:02
Lattice Regression - 311:15
Effect of Different Lattice Regression Regularizers - 111:54
Effect of Different Lattice Regression Regularizers - 212:45
Lattice Regression Closed Form Solution13:52
Example Color Management Results - 115:08
Gaussian Process Regression17:48
Example Color Management Results - 219:04
Omnidirectional Super-resolution19:55
Omnidirectional Superres Related Work20:39
Lattice Regression Approach - 121:08
Lattice Regression Approach - 222:33
Lattice Regression Approach - 323:21
Lattice Regression Approach - 424:16
Visual Homing25:16
Some Conclusions - 126:36
Some Conclusions - 226:46
Some Conclusions - 326:57
Some Conclusions - 427:11
For details, see ...27:22
Inverse Device Characterization - 130:43