Guided Mesh Normal Filtering Wangyu Zhang USTC Bailin Deng EPFL, - - PowerPoint PPT Presentation

▶

Jan 06, 2023 106 likes •385 views

Guided Mesh Normal Filtering Wangyu Zhang USTC Bailin Deng EPFL, University of Hull Juyong Zhang USTC Sofien Bouaziz EPFL Ligang Liu USTC Filtering is necessary Filtering Related Work Non-iterative, feature-preserving mesh

SLIDE 1

Guided ¡Mesh ¡Normal ¡Filtering

Wangyu Zhang Bailin Deng Juyong Zhang Sofien Bouaziz Ligang Liu USTC EPFL, University of Hull USTC EPFL USTC

SLIDE 2

Filtering is necessary

Filtering

SLIDE 3

Related Work

Non-iterative, feature-preserving mesh smoothing [Jones et al. 2003] Bilateral mesh denoising [Fleishman et al. 2003] Bilateral Normal Filtering for Mesh Denoising [Zheng et al. 2011] Mesh Denoising via L0 Minimization [He & Schaefer 2013]

SLIDE 4

Bilateral filter

SLIDE 5

Joint bilateral filter

SLIDE 6

SLIDE 7

The role of guidance

The success of joint bilateral filtering is heavily

dependent on the guidance signal.

The guidance signal should provide a robust

estimation about the features of the output signal

SLIDE 8

The importance of guidance

Original Noisy With Ground Truth Normals With Noisy Normals

SLIDE 9

Guidance geometry

Contrary to the case of images, such guidance

geometry is not easily available from measure devices.

It often has to be constructed computationally.

SLIDE 10

Example

Noisy With Ground Truth Normals With Noisy Normals Ours

SLIDE 11

Denoising pipeline

!"#$%&'()($#*+, !"#$%&'()($#*+, !"#$%&'()($#*+,

*.,")
*.,")
*.,")

/0*1",$( /0*1",$( /0*1",$( 2,30# 40#30#

567 67

SLIDE 12

Guidance normal computation

SLIDE 13

Normal consistency

SLIDE 14

Maximum normal difference

SLIDE 15

Edge saliency measurement

SLIDE 16

Patch selection

SLIDE 17

Normal filtering

Our normal filtering computes a new unit normal

for each face via joint bilateral filter:

SLIDE 18

Updating vertices

Based on the filtering face normals, the vertex

positions are updated by minimizing the error of the compatibility conditions:

SLIDE 19

Recap: pipeline

!"#$%&'()($#*+, !"#$%&'()($#*+, !"#$%&'()($#*+,

*.,")
*.,")
*.,")

/0*1",$( /0*1",$( /0*1",$( 2,30# 40#30#

567 67

SLIDE 20

Denoising process

SLIDE 21

Results: vs bilateral filter

Ours Input

[ZFAT11]

SLIDE 22

Results: comparisons

!"#$%&&'()*+,-*. !"#$%&&'()/*+0-*. !12&3' !4567&8' 9:;< !"#$%&'( !)##&'( !*+,-&.( !/0"122(34567859

!"#$%

SLIDE 23

Time statistics

!"#$% &'$()*+$, &-.+$, /*0$1,234)$(

.5#*,6

789: ;<=87 >?>97 @%"+6 A99; ;9::> >?;>8 @B55C D8AD8 7=8:; >?7=A 4("5 A::98 ;7A<A: ;?:9;

SLIDE 24

Conclusion

A joint bilateral filter for mesh processing

! A novel method to construct the guidance signal ! Effective and efficient, simple to implement ! Much better denoising results than state of the art

SLIDE 25

$%&'()*+,&',)-&'.('/0/.1/21&3''

%445(3667.4%*28,)9621-&:76;*.-&-<&:).(.:7

SLIDE 26