Flash Photography Enhancement via Intrinsic Relighting

Similar documents
Computational Illumination Frédo Durand MIT - EECS

Agenda. Fusion and Reconstruction. Image Fusion & Reconstruction. Image Fusion & Reconstruction. Dr. Yossi Rubner.

Flash Photography Enhancement via Intrinsic Relighting

Flash Photography Enhancement via Intrinsic Relighting

A Gentle Introduction to Bilateral Filtering and its Applications 08/10: Applications: Advanced uses of Bilateral Filters

Fixing the Gaussian Blur : the Bilateral Filter

Computational Photography

A Gentle Introduction to Bilateral Filtering and its Applications 08/10: Applications: Advanced uses of Bilateral Filters

Applications of Flash and No-Flash Image Pairs in Mobile Phone Photography

Fast Bilateral Filtering for the Display of High-Dynamic-Range Images

Prof. Feng Liu. Spring /12/2017

Computational Photography and Video. Prof. Marc Pollefeys

Problem Set 3. Assigned: March 9, 2006 Due: March 23, (Optional) Multiple-Exposure HDR Images

! High&Dynamic!Range!Imaging! Slides!from!Marc!Pollefeys,!Gabriel! Brostow!(and!Alyosha!Efros!and! others)!!

Image Enhancement of Low-light Scenes with Near-infrared Flash Images

High dynamic range imaging and tonemapping

Image Enhancement of Low-light Scenes with Near-infrared Flash Images

Comp Computational Photography Spatially Varying White Balance. Megha Pandey. Sept. 16, 2008

Computational Illumination

Limitations of the Medium, compensation or accentuation

Limitations of the medium

Tonemapping and bilateral filtering

Guided Filtering Using Reflected IR Image for Improving Quality of Depth Image

Fast Bilateral Filtering for the Display of High-Dynamic-Range Images

Burst Photography! EE367/CS448I: Computational Imaging and Display! stanford.edu/class/ee367! Lecture 7! Gordon Wetzstein! Stanford University!

Denoising and Effective Contrast Enhancement for Dynamic Range Mapping

CSC 320 H1S CSC320 Exam Study Guide (Last updated: April 2, 2015) Winter 2015

Multispectral Image Dense Matching

Digital Image Processing

High Dynamic Range Video with Ghost Removal

Prof. Feng Liu. Winter /10/2019

Tone mapping. Digital Visual Effects, Spring 2009 Yung-Yu Chuang. with slides by Fredo Durand, and Alexei Efros

Automatic Content-aware Non-Photorealistic Rendering of Images

Realistic Image Synthesis

Art Photographic Detail Enhancement

Tone Adjustment of Underexposed Images Using Dynamic Range Remapping

Deblurring. Basics, Problem definition and variants

Camera Exposure Modes

A Novel Hybrid Exposure Fusion Using Boosting Laplacian Pyramid

6.A44 Computational Photography

APPLICATION OF PATTERNS TO IMAGE FEATURES

Computational 4/23/2009. Computational Illumination: SIGGRAPH 2006 Course. Course WebPage: Flash Shutter Open

Image Visibility Restoration Using Fast-Weighted Guided Image Filter

The Dynamic Range Problem. High Dynamic Range (HDR) Multiple Exposure Photography. Multiple Exposure Photography. Dr. Yossi Rubner.

Implementation of Image Deblurring Techniques in Java

High Dynamic Range Images : Rendering and Image Processing Alexei Efros. The Grandma Problem

The Unique Role of Lucis Differential Hysteresis Processing (DHP) in Digital Image Enhancement

Limitations of the Medium, compensation or accentuation: Contrast & Palette

Dynamic Range. H. David Stein

Preserving Natural Scene Lighting by Strobe-lit Video

Multispectral Bilateral Video Fusion

When you first open the dialog box you only see two sliders.

Local Adjustment Tools

Achim J. Lilienthal Mobile Robotics and Olfaction Lab, AASS, Örebro University

Ian Barber Photography

Automatic Selection of Brackets for HDR Image Creation

Image Deblurring with Blurred/Noisy Image Pairs

Tablet overrides: overrides current settings for opacity and size based on pen pressure.

Guided Image Filtering for Image Enhancement

Maine Day in May. 54 Chapter 2: Painterly Techniques for Non-Painters

Computational Camera & Photography: Coded Imaging

Making better photos. Better Photos. Today s Agenda. Today s Agenda. What makes a good picture?! Tone Style Enhancement! What makes a good picture?!

Continuous Flash. October 1, Technical Report MSR-TR Microsoft Research Microsoft Corporation One Microsoft Way Redmond, WA 98052

High-Dynamic-Range Imaging & Tone Mapping

Photo Editing Workflow

icam06, HDR, and Image Appearance

Contrast Image Correction Method

6.098 Digital and Computational Photography Advanced Computational Photography. Bill Freeman Frédo Durand MIT - EECS

More image filtering , , Computational Photography Fall 2017, Lecture 4

Dappled Photography: Mask Enhanced Cameras for Heterodyned Light Fields and Coded Aperture Refocusing

1. Any wide view of a physical space. a. Panorama c. Landscape e. Panning b. Grayscale d. Aperture

CS354 Computer Graphics Computational Photography. Qixing Huang April 23 th 2018

Color Reproduction. Chapter 6

PHOTOGRAPHY: MINI-SYMPOSIUM

Selective Edits in Camera Raw

Image compression using sparse colour sampling combined with nonlinear image processing

Neuron Bundle 12: Digital Film Tools

Two-scale Tone Management for Photographic Look

Motion Estimation from a Single Blurred Image

prepared by Allison Hwang for T. Purdy 2011

Creating a Contrast Mask. Text and images Copyright (C) 2002 Eric R. Jeschke and may not be used without permission of the author.

Computational Photography

Anti-shaking Algorithm for the Mobile Phone Camera in Dim Light Conditions

Filtering Images in the Spatial Domain Chapter 3b G&W. Ross Whitaker (modified by Guido Gerig) School of Computing University of Utah

POLAROID EMULATION INCREASED CONTRAST, SATURATION & CLARITY

1. Any wide view of a physical space. a. Panorama c. Landscape e. Panning b. Grayscale d. Aperture


Correcting Over-Exposure in Photographs

Coded photography , , Computational Photography Fall 2018, Lecture 14

Using VLSI for Full-HD Video/frames Double Integral Image Architecture Design of Guided Filter

A collection of example photos SB-900

Camera controls. Aperture Priority, Shutter Priority & Manual

How to Create Fake Shadows

Gradient-Based Correction of Chromatic Aberration in the Joint Acquisition of Color and Near-Infrared Images

Image Processing. Adam Finkelstein Princeton University COS 426, Spring 2019

Visual Perception of Images

Basic Digital Dark Room

CHAPTER 12 - HIGH DYNAMIC RANGE IMAGES

CSE 564: Scientific Visualization

PHIL MORGAN PHOTOGRAPHY

Transcription:

Flash Photography Enhancement via Intrinsic Relighting Elmar Eisemann MIT/Artis-INRIA Frédo Durand MIT

Introduction Satisfactory photos in dark environments are challenging!

Introduction Available light: + nice lighting - noise/blurriness - color No-flash

Introduction Flash: + details + color - flat/artificial - flash shadows - red eyes Flash

Introduction Our approach: Use no-flash image relight flash image No-flash Flash Result

Introduction Our approach: Use no-flash image relight flash image No-flash + original lighting + details/sharpness + color Result

Introduction One approach: Blend the two photos No-flash + Flash Blending

Introduction One approach: Blend the two photos Blending Our result

Introduction One approach: Blend the two photos Blending Our solution: more details and less noise Our result

Overview Related Work Our Approach Results Conclusion and Future Work

Overview Related Work Our Approach Results Conclusion and Future Work

Related Work Relighting: Decouple luminance and texture Barrow and Tennenbaum [1978] Weiss [2001] using image sequences Tappen et al. [2003] from a single image Oh et al. [2001] Input Image Luminance Texture Images courtesy of Oh et al.

Related Work Tone Mapping of High Dynamic Range Images Decouple detail / large-scale information Tumblin et al. [1999] Durand et al. [2002] Choudhury et al. [2003] + Input Detail Large-Scale Output

Related Work Jia et al. [2004]: Solve problem of blur in long exposures Image pair with different exposure times Short exposure Color manipulation Result Long exposure Details from short exposure Images courtesy of Jia et al.

Related Work Petschnigg et al.[2004]: many similarities previous talk discussion at the end

Overview Related Work Our Approach Results Conclusion and Future Work

Our Approach - Main Idea

Our Approach

Our Approach

Our Approach Registration Registration

Registration Compensate for camera motion (image translation) Difficult because lighting changes Edge detection No-flash See also Ward[2004], Kang[2003] Flash

Our Approach Registration Registration

Our Approach Our Approach Decomposition

Decomposition Color / Intensity: = * original intensity color

Our Approach Our Approach Decomposition

Our Approach Decoupling

Decoupling Lighting : Large-scale variation Texture : Small-scale variation Lighting Texture

Large-scale Layer Bilateral filter edge preserving filter Smith and Brady 1997; Tomasi and Manducci 1998; Durand et al. 2002 Input Output

Large-scale Layer Bilateral filter

Cross Bilateral Filter Similar to joint bilateral filter by Petschnigg et al. When no-flash image is too noisy Borrow similarity from flash image edge stopping from flash image See detail in paper Bilateral Cross Bilateral

Detail Layer / = Intensity Large-scale Detail Recombination: Large scale * Detail = Intensity

Recombination * = Large-scale No-flash Detail Flash Intensity Result Recombination: Large scale * Detail = Intensity

Recombination shadows ~ * ~ Intensity Result Color Flash Result Recombination: Intensity * Color = Original

Our Approach

Our Approach Shadow Detection/Treatment

Shadow Correction Why? No flash Flash Global Straightforward white balance recombination shadows

Shadow Correction Why? Global white balance in shadows Several artifacts: at shadow boundary inside shadows (color bleeding) shadows need to be corrected!... and detected

Shadow Detection Flash Umbra Penumbra Detection in two steps

Shadow Detection Umbra detection No direct light from flash Difference of the two photos Ι reveals these regions - = Flash No-flash Ι

Shadow Detection Umbra detection Difference Ι = light from the flash? Goal: Find a threshold for Ι Automatic Threshold Detection (see paper) Ι

Shadow Detection Penumbra detection strong gradient at boundary no strong gradient in no-flash image connected to umbra No-flash flash Umbra Penumbra

Shadow Correction Correct color and detail Shadow mask

Shadow Color Correction Correct color Wrong color Flash color Fill in shadow from similar surrounding No-flash colors

Shadow Color Correction No-flash Flash

Shadow Color Correction No-flash Flash

Shadow Color Correction No-flash Flash Outside shadow

Shadow Color Correction No-flash Flash Inside shadow Outside shadow

Shadow Color Correction No-flash Flash Inside shadow Outside shadow Select pixel in shadow

Shadow Color Correction No-flash Flash Corresponding pixel Inside shadow Outside shadow

Shadow Color Correction No-flash Flash Spatial weights Inside shadow Outside shadow

Shadow Color Correction No-flash Flash Spatial and Color weights Inside shadow Outside shadow

Shadow Color Correction No-flash Flash Spatial and Color weights Inside shadow Outside shadow

Shadow Color Correction No-flash Flash Use shadow mask Inside shadow Outside shadow

Shadow Color Correction No-flash Flash Use shadow mask Inside shadow Outside shadow

Shadow Color Correction No-flash Flash Use shadow mask Inside shadow Outside shadow

Shadow Color Correction No-flash Flash Use shadow mask Inside shadow Outside shadow

Shadow Color Correction No-flash Flash Use shadow mask Inside shadow Outside shadow

Shadow Color Correction No-flash Flash Use shadow mask Inside shadow Outside shadow

Shadow Color Correction No-flash Flash Use weights on flash color Inside shadow Outside shadow

Shadow Color Correction No-flash Flash Replace shadow pixel Inside shadow Outside shadow

Shadow Color Correction No-flash Flash Inside shadow Proceed for all shadow pixels Outside shadow

Our Approach

Our Approach

Overview Related Work Our Approach Results Conclusion and Future Work

Results No-flash Flash

Results No correction Our result

Results No-flash Flash

Results No-flash Flash

Results No-flash Flash Our result

Results Our result

Results Our result

Emphasize Foreground Our result Deduce distance to camera Exploit 1/r 2 flash intensity falloff (see paper) Emphasized foreground

(Inverse) White Balance No-flash Flash Retain warm tones from available lighting (see paper)

Results No-flash Flash Result

Overview Related Work Our Approach Results Conclusion and Future Work

Conclusion Improving photography in dim environments Capture original lighting Add sharpness/details Cross bilateral filter Correct flash shadows No-flash Result Pseudo distance (emphasize foreground) (Inverse) white balance

Future Work Local coherence for shadow detection Different recombinations Bilateral filter parameters Infra-red flash

Aknowledgements École Normale Supérieure - Paris Deshpande Center MIT-France NSF CISE

Thank you very much for your attention!

Comparison: Similarities Petschnigg et al. Eisemann et al. Detail transfer or relighting Use bilateral filter Joint or cross bilateral filter

Comparison: Color Petschnigg et al. Eisemann et al. Color from no-flash More noisy Less need for shadow correction White balance no-flash image Color from flash Less noisy Need correction in shadow Inverse white balance flash image

Comparison: Differences Petschnigg et al. Eisemann et al. Color from no-flash Semi-manual shadows/highlights Continuous flash Red-eye removal Algorithmic differences Additional tools Color from flash Automatic shadows Pseudo distance, emphasize foreground

Thank you again for your attention!