A Computational Darkroom for BW Photography.


207 views
Uploaded on:
Description
A Computational Darkroom for BW Photography. Soonmin Bae, Sylvain Paris, and Fr é do Durand Current Status : Resubmission to Siggraph. Goals. To improve highly contrasting photos "Look" exchange between two pictures Direct interjection and control of the "look".
Transcripts
Slide 1

A Computational Darkroom for BW Photography Soonmin Bae, Sylvain Paris, and Fr é do Durand Current Status : Resubmission to Siggraph

Slide 2

Objectives To improve highly contrasting photos “Look” exchange between two pictures Direct addition and control of the “look”

Slide 3

What is the issue? Direct change to B&W yields frequently uninspiring results.

Slide 4

What can be the “Look”

Slide 5

Approaches Decomposition of a picture into substantial scale variety layer and high recurrence surface layer Control the worldwide difference and the nearby textureness independently Quantitative portrayal Use picture measurements and histograms

Slide 6

What we point at… Control of the visual quality, “look” Parametric portrayal User-situated and natural technique HDR pictures

Slide 7

What we don\'t do… Deal with Color photos Paintings Change Content Change Composition Crop Select a model or perfect parameters

Slide 8

What they do versus What we do Objective tone generation versus Control of the look Non-parametric versus Parametric portrayal Tone mapping Ferwerda et al. 1996;Tumblin and Rushmeier 1993; Ward 1994 Ashikhmin 2002; Tumblin and Turk 1999; Pattanaik et al. 1998; Reinhard et al. 2002 Color2gray Gooch et al. 2005 Image analogies Hertzmann et al. 2001; Efros and Freeman 2001; Rosales et al. 2003; Drori et al. 2003

Slide 9

Challenges Identification of vital visual attributes Meaningful component choice Decomposition Faithful extraction of the elements Reconstruction Visual antique (basically radiances) Subjective issues Preference versus Comparability

Slide 10

Quick Technical Overview expansive scale Challenge: separate composition from edges. “textureness” information point of interest

Slide 11

Quick Technical Overview Histogram control (exchange conceivable)

Slide 12

Quick Technical Overview Histogram control of the “textureness”

Slide 13

Quick Technical Overview after before

Slide 14

Exploring Various Options in a Few Clicks

Slide 15

Preliminary Results Model Input

Slide 16

Open Discussion Should we incorporate the accompanying spaces? Shading photos Paintings Which ought to be sought after? Exchange versus Direct parameter change Similarity ve

Recommended
View more...