OPTIMIZED SAMPLING FOR VIEW INTERPOLATION IN LIGHT FIELDS WITH OVERLAPPING PATCHES

Similar documents
Coded Exposure HDR Light-Field Video Recording

IMAGE RESTORATION WITH NEURAL NETWORKS. Orazio Gallo Work with Hang Zhao, Iuri Frosio, Jan Kautz

Modeling the calibration pipeline of the Lytro camera for high quality light-field image reconstruction

Liu Yang, Bong-Joo Jang, Sanghun Lim, Ki-Chang Kwon, Suk-Hwan Lee, Ki-Ryong Kwon 1. INTRODUCTION

Super resolution with Epitomes

360 Panorama Super-resolution using Deep Convolutional Networks

DEEP LEARNING ON RF DATA. Adam Thompson Senior Solutions Architect March 29, 2018

LEGO car course topics

Image Enhancement for Astronomical Scenes. Jacob Lucas The Boeing Company Brandoch Calef The Boeing Company Keith Knox Air Force Research Laboratory

Dictionary Learning based Color Demosaicing for Plenoptic Cameras

Compressive Coded Aperture Superresolution Image Reconstruction

Filters. Materials from Prof. Klaus Mueller

Coded Aperture for Projector and Camera for Robust 3D measurement

Optimized Quality and Structure Using Adaptive Total Variation and MM Algorithm for Single Image Super-Resolution

IIL Imaging Model, Grating-Based Analysis and Optimization

Image Enhancement for Astronomical Scenes. Jacob Lucas The Boeing Company Brandoch Calef The Boeing Company Keith Knox Air Force Research Laboratory

fast blur removal for wearable QR code scanners

DYNAMIC CONVOLUTIONAL NEURAL NETWORK FOR IMAGE SUPER- RESOLUTION

Computer Vision. Howie Choset Introduction to Robotics

Sampling Rate = Resolution Quantization Level = Color Depth = Bit Depth = Number of Colors

CP-JKU SUBMISSIONS FOR DCASE-2016: A HYBRID APPROACH USING BINAURAL I-VECTORS AND DEEP CONVOLUTIONAL NEURAL NETWORKS

Efficient Diversity Technique for Hybrid Narrowband-Powerline/Wireless Smart Grid Communications

Hot Electron Injection Field Effect Transistor

Modeling and Synthesis of Aperture Effects in Cameras

Dynamic Scene Deblurring Using Spatially Variant Recurrent Neural Networks

An evaluation of debayering algorithms on GPU for real-time panoramic video recording

multiframe visual-inertial blur estimation and removal for unmodified smartphones

Computational Sensors

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

Measurement of Texture Loss for JPEG 2000 Compression Peter D. Burns and Don Williams* Burns Digital Imaging and *Image Science Associates

arxiv: v1 [physics.data-an] 3 Mar 2016

Compressive Wireless Pulse Sensing

Image Processing & Projective geometry

Random Coded Sampling for High-Speed HDR Video

Coding & Signal Processing for Holographic Data Storage. Vijayakumar Bhagavatula

Continuously Variable Bandwidth Sharp FIR Filters with Low Complexity

Snapshot Hyperspectral Light Field Imaging

LWIR NUC Using an Uncooled Microbolometer Camera

High Dynamic Range Imaging

Compressive Imaging. Aswin Sankaranarayanan (Computational Photography Fall 2017)

Compressive Through-focus Imaging

Double resolution from a set of aliased images

Thermal Image Enhancement Using Convolutional Neural Network

Multispectral Image Dense Matching

Learning Sensor Multiplexing Design through Back-propagation

1. Describe how a graphic would be stored in memory using a bit-mapped graphics package.

A Novel Differential Microstrip Patch Antenna and Array at 79 GHz #Z. Tong 1, A. Stelzer 1,2, C. Wagner 2, R. Feger 2, E.

Create a 3D Model from Photos with Free and Open Source Software (FOSS)

Space-Time-Brightness Sampling Using an Adaptive Pixel-Wise Coded Exposure

Short-course Compressive Sensing of Videos

Coding and Modulation in Cameras

Components for virtual environments Michael Haller, Roland Holm, Markus Priglinger, Jens Volkert, and Roland Wagner Johannes Kepler University of Linz

Audio Imputation Using the Non-negative Hidden Markov Model

DodgeCmd Image Dodging Algorithm A Technical White Paper

image Scanner, digital camera, media, brushes,

Lecture 5: Sinusoidal Modeling

VQ Source Models: Perceptual & Phase Issues

ESE 531: Digital Signal Processing

High Fidelity 3D Reconstruction

Project Title: Sparse Image Reconstruction with Trainable Image priors

Computational Approaches to Cameras


Image Scaling. This image is too big to fit on the screen. How can we reduce it? How to generate a halfsized

Reverse Engineering the Human Vision System

Convolution Pyramids. Zeev Farbman, Raanan Fattal and Dani Lischinski SIGGRAPH Asia Conference (2011) Julian Steil. Prof. Dr.

Color Imaging Conference 16, Portland OR 11/14/2008

Creating Intelligence at the Edge

Sampling and Pyramids

Supplementary Material of

16nm with 193nm Immersion Lithography and Double Exposure

RF Radar Systems. C. G. Diskus 1, A. Stelzer 2. Altenberger Straße 69, 4040 Linz, Austria

Course Overview. Dr. Edmund Lam. Department of Electrical and Electronic Engineering The University of Hong Kong

Performance Evaluation of H.264 AVC Using CABAC Entropy Coding For Image Compression

CS448f: Image Processing For Photography and Vision. Fast Filtering Continued

High Resolution Spectral Video Capture & Computational Photography Xun Cao ( 曹汛 )

A survey of Super resolution Techniques

Learning Pixel-Distribution Prior with Wider Convolution for Image Denoising

Light-Field Microscopy: A Review

Research on Hand Gesture Recognition Using Convolutional Neural Network

Depth Estimation Algorithm for Color Coded Aperture Camera

Aliasing and Antialiasing. What is Aliasing? What is Aliasing? What is Aliasing?

arxiv: v1 [cs.cv] 24 Nov 2017

Antialiasing and Related Issues

Image Based Subpixel Techniques for Movement and Vibration Tracking

Computational Efficiency of the GF and the RMF Transforms for Quaternary Logic Functions on CPUs and GPUs

General MIMO Framework for Multipath Exploitation in Through-the-Wall Radar Imaging

Blind Single-Image Super Resolution Reconstruction with Defocus Blur

Dual-fisheye Lens Stitching for 360-degree Imaging & Video. Tuan Ho, PhD. Student Electrical Engineering Dept., UT Arlington

Chapter IV THEORY OF CELP CODING

Ben Baker. Sponsored by:

Constrained Channel Estimation Methods in Underwater Acoustics

Infocommunication. Sampling, Quantization. - Bálint TÓTH, BME TMIT -

Technical Specifications: tog VR

Computational Photography

Analysis and Synthesis of Texture

Pairwise Optimization of Modulation Constellations for Non-Uniform Sources

Blind Correction of Optical Aberrations

신경망기반자동번역기술. Konkuk University Computational Intelligence Lab. 김강일

PARALLEL ALGORITHMS FOR HISTOGRAM-BASED IMAGE REGISTRATION. Benjamin Guthier, Stephan Kopf, Matthias Wichtlhuber, Wolfgang Effelsberg

A Single Image Haze Removal Algorithm Using Color Attenuation Prior

Transcription:

OPTIMIZED SAMPLING FOR VIEW INTERPOLATION IN LIGHT FIELDS WITH OVERLAPPING PATCHES D. C. Schedl 1 and O. Bimber 1 1 Institute of Computer Graphics, david.schedl@jku.at and oliver.bimber@jku.at

MOTIVATION dense (225 samples) uniform sparse (64 samples) [ours] coded sparse (225 from 64 samples) D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 2

RELATED WORK Coded sampling with local dictionaries [Schedl et al., ICCP 2015] [Schedl et al., CVIU 2017] D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 3

RELATED WORK Coded sampling with local dictionaries Compressed Sensing [Marwah et al., TOG 2013] [Cao et al., Opt. Express 2014] D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 4

RELATED WORK Coded sampling with local dictionaries Compressed Sensing Depth-based view interpolation Learning-based methods [Kalantari et al., TOG 2016] D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 5

RELATED WORK Coded sampling with local dictionaries Compressed Sensing Depth-based view interpolation Learning-based methods Other [Shi et al., TOG 2014] [Vagharshakyan et al., PAMI 2015] D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 6

CONTRIBUTIONS New sampling quality metric A reduced search space for sampling mask estimation An enhanced upsampling technique supporting maximal patch overlaps D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 7

minimum distance SAMPLING QUALITY METRIC D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 8

minimum distance SAMPLING QUALITY METRIC interpolation extrap. D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 9

ours SAMPLING QUALITY METRIC interpolation extrap. D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 10

SAMPLING QUALITY METRIC D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 11

SAMPLING QUALITY METRIC sampling mask D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 12

SAMPLING PATTERN ESTIMATION Constraints: regular symmetric sampling mask (64 samples; 15x15 grid) D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 13

SAMPLING PATTERN ESTIMATION basis grid permutations sampling mask (only guidance) D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 14

SAMPLING PATTERN ESTIMATION E: 0.188 0.184 0.191 basis grid permutations D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 15

SAMPLING PATTERN ESTIMATION E: 0.188 0.184 0.191 basis grid permutations final sampling mask D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 16

RECONSTRUCTION sampling mask dictionary D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 17

RECONSTRUCTION sub-sampled light-field sampling mask reconstructed light-field D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 18

RECONSTRUCTION sampling mask D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 19

RECONSTRUCTION reconstructed light-field D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 20

Schedl 15 Schedl 17 Ours RESULTS: SAMPLING MASKS 64 72 69 48 high (bad) low (good) unsupported unsupported unsupported D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 21

RESULTS Scenes (N) Marwah '13 Shi '14 Schedl '15 Kalantari '16 Schedl '17 Ours Table Amethyst (64) 37.77dB - - 40.11dB 41.86dB 42.08dB Lego (64) 28.79dB - - 32.87dB 35.63dB 37.26dB Lego (48) - - - - 33.86dB 35.75dB Cave (64) 26.51dB - - 30.99dB 38.57dB 41.08dB Alley (64) 36.58dB - - 43.23dB 43.83dB 44.35dB Amethyst (72) - 36.40dB - - 42.18dB 42.55dB Tarot (72) - 30.19dB - - 37.81dB 39.20dB Amethyst (69) - - 41.91dB - 42.07dB 42.43dB Tarot (69) - - 34.09dB - 37.88dB 39.04dB Tarot (48) - - - - 35.96dB 37.54dB Cave (69) - - 29.96dB - 39.14dB 41.41dB Alley (69) - - 41.36dB - 44.24dB 45.20dB D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 22

Lego: Schedl 17 (64) Ours (64) Reference (225) D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 23

Cave: Kalantari 16 (64) Ours (64) Reference (225) D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 24

Tarot: Schedl 15 (69) Ours (69) Reference (225) D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 25

Amethyst: Shi '14 (72) Ours (72) Reference (225) D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 26

Alley: Marwah 13 (64) Ours (64) Reference (225) D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 27

LIMITATIONS / FUTURE WORK Time: 40h 5 days on NVIDIA Tesla V100 GPU D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 29

LIMITATIONS / FUTURE WORK Time: 40h 5 days on NVIDIA Tesla V100 GPU Other light-field camera designs D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 30

LIMITATIONS / FUTURE WORK Time: 40h 5 days on NVIDIA Tesla V100 GPU Other light-field camera designs Other fields (e.g. image-based relighting) D. C. Schedl & O. Bimber / Optimized Sampling for View Interpolation in Light Fields with Overlapping Patches 31

Schedl 17 (64) Ours (64) Reference (225) Kalantari 16 (64) Ours (64) Reference (225) More information: www.jku.at/cg Contact: david.schedl@jku.at and oliver.bimber@jku.at This project was funded by FWF (P 28581-N33) JOHANNES KEPLER UNIVERSITY LINZ Altenberger Str. 69 4040 Linz, Austria www.jku.at