SlideShare a Scribd company logo
1 of 46
Download to read offline
Dense Variational Reconstruction of Non-Rigid
Surfaces from Monocular Video
Ravi Garg Anastasios Roussos∗
Lourdes Agapito∗
Queen Mary, University of London
∗
Now at UCL
Before This Paper
1 / 17
Dense Non Rigid Structure from Motion
Input: monocular sequence of non-rigid surface.
...
Goal: dense 3D reconstruction for every frame.
2 / 17
Dense Non Rigid Structure from Motion
Input: monocular sequence of non-rigid surface.
...
Goal: dense 3D reconstruction for every frame.
2 / 17
Dense Non Rigid Structure from Motion
Input: monocular sequence of non-rigid surface.
...
Goal: dense 3D reconstruction for every frame.
2 / 17
Dense Non Rigid Structure from Motion
Input: monocular sequence of non-rigid surface.
...
Goal: dense 3D reconstruction for every frame.
2 / 17
Dense Non Rigid Structure from Motion
Input: monocular sequence of non-rigid surface.
...
Goal: dense 3D reconstruction for every frame.
2 / 17
Dense Non Rigid Structure from Motion
Input: monocular sequence of non-rigid surface.
...
Goal: dense 3D reconstruction for every frame.
Goal: dense 3D reconstruction for every frame.
NO additional sensors.
NO pre-trained shape models.
NO surface template.
2 / 17
Dense Non Rigid Structure from Motion
Input: monocular sequence of non-rigid surface.
...
NRSfM: ill posed problem
Goal: dense 3D reconstruction for every frame.
Goal: dense 3D reconstruction for every frame.
NO additional sensors.
NO pre-trained shape models.
NO surface template.
2 / 17
Traditional Sparse Non Rigid Structure from Motion
...
3 / 17
Traditional Sparse Non Rigid Structure from Motion
...Feature
Tracking
...
3 / 17
Traditional Sparse Non Rigid Structure from Motion
...Feature
Tracking
...
...3D Shape
Inference
3 / 17
Traditional Sparse Non Rigid Structure from Motion
Priors
...Feature
Tracking
... +
...3D Shape
Inference
3 / 17
Low Rank Prior for NRSfM
Shape Space
(Bregler, Hertzmann, Biermann, Recovering non-rigid 3D shape from image streams CVPR’00.)
Bregler et al. CVPR’00, Brand CVPR’01, Xiao et al. IJCV’06, Torresani et al. PAMI’08, Akhter et al.
CVPR’09, Bartoli et al. CVPR2008, Paladini et al. IJCV’12,Dai et al. CVPR’12...
4 / 17
Low Rank Prior for NRSfM
Shape Space
(Bregler, Hertzmann, Biermann, Recovering non-rigid 3D shape from image streams CVPR’00.)
(Park et al. ECCV’10)
4 / 17
Inspiration from Dense Rigid Reconstruction
(Newcombe, Lovegrove, Davison, DTAM: Dense Tracking and Mapping in Real-Time, ICCV’11)
5 / 17
Inspiration from Dense Rigid Reconstruction
(Newcombe, Lovegrove, Davison, DTAM: Dense Tracking and Mapping in Real-Time, ICCV’11)
Key features
Variational approach.
Use of smoothness priors.
Per pixel reconstruction.
Scalable and GPU friendly algorithm.
5 / 17
Leap from sparse to dense NRSfM
Sparse
Dai et al. CVPR’12
Dense
This work
6 / 17
Leap from sparse to dense NRSfM
Sparse
Dai et al. CVPR’12
Dense
This work
We take the best of both worlds:
Low rank prior from sparse non rigid SfM.
Variational framework from dense rigid SfM.
6 / 17
Leap from sparse to dense NRSfM
Sparse
Dai et al. CVPR’12
Dense
This work
We take the best of both worlds:
Low rank prior from sparse non rigid SfM.
Variational framework from dense rigid SfM.
Our contribution
First variational formulation to dense NRSfM.
Scalable algorithm which can be ported on GPU.
6 / 17
Our Approach in a Nutshell
...
7 / 17
Our Approach in a Nutshell
...
...
Step 1: Dense
Video Registration ...
7 / 17
Our Approach in a Nutshell
Step 2: Dense
Shape Inference
...
...
...
Step 1: Dense
Video Registration ...
7 / 17
Our Approach in a Nutshell
Step 2: Dense
Shape Inference
...
...
...
Step 1: Dense
Video Registration ...
Priors+
7 / 17
Our Approach in a Nutshell
Step 2: Dense
Shape Inference
...
...
...
Step 1: Dense
Video Registration ...
Low rank.
Spatial smoothness.+
7 / 17
Our Approach in a Nutshell
...
... Low rank.
Spatial smoothness.
Step 1: Dense
Video Registration ...
+
Garg, Roussos, Agapito, A variational approach to video registration with subspace constraints, IJCV’13.
7 / 17
Our Approach in a Nutshell
Step 2: Dense
Shape Inference
...
... Low rank.
Spatial smoothness.+
7 / 17
Orthographic Projection Model
8 / 17
Orthographic Projection Model
8 / 17
Orthographic Projection Model
8 / 17
Orthographic Projection Model
8 / 17
Orthographic Projection Model
W = RS
8 / 17
Energy Minimisation Approach to NRSfM
Formulation of a single unified energy to estimate:
Orthographic projection matrices
3D shapes for all the frames
E R , S = λ Edata R, S + Ereg S + τ Etrace S
reprojection error over all frames
spatial smoothness prior on 3D shapes
low rank prior on 3D shapes
9 / 17
Energy Minimisation Approach to NRSfM
Formulation of a single unified energy to estimate:
Orthographic projection matrices
3D shapes for all the frames
E R , S = λ Edata R, S + Ereg S + τ Etrace S
reprojection error over all frames
spatial smoothness prior on 3D shapes
low rank prior on 3D shapes
9 / 17
Energy Minimisation Approach to NRSfM
Formulation of a single unified energy to estimate:
Orthographic projection matrices
3D shapes for all the frames
E R , S = λ Edata R, S + Ereg S + τ Etrace S
reprojection error over all frames
spatial smoothness prior on 3D shapes
low rank prior on 3D shapes
9 / 17
Energy Minimisation Approach to NRSfM
Formulation of a single unified energy to estimate:
Orthographic projection matrices
3D shapes for all the frames
E R , S = λ Edata R, S + Ereg S + τ Etrace S
reprojection error over all frames
spatial smoothness prior on 3D shapes
low rank prior on 3D shapes
9 / 17
Reprojection Error
E
`
R, S
´
= λEdata
`
R, S
´
+ Ereg
`
S
´
+ τEtrace
`
S
´
Edata (R, S) = W − RS 2
F
10 / 17
Spatial Smoothness Prior
E
`
R, S
´
= λEdata
`
R, S
´
+ Ereg
`
S
´
+ τEtrace
`
S
´
Ereg S =
i
TV (Si)
−−−−−−→
Without regularisation With regularisation
11 / 17
Low Rank Prior
E
`
R, S
´
= λEdata
`
R, S
´
+ Ereg
`
S
´
+ τEtrace
`
S
´
Etrace S = S ∗ =
i
σi(S)
lies in−−−−→ span
K F
Angst et al. ECCV’12, Dai et al. CVPR’12, Angst et al. ICCV’11, Dai et al. ECCV’10
12 / 17
Minimisation of E R, S
min
R,S
λ W − RS 2
F
Reprojection
error
+
i
TV (Si)
Smoothness
prior
+ τ S ∗
Low rank
prior
13 / 17
Minimisation of E R, S
min
R,S
λ W − RS 2
F
Reprojection
error
+
i
TV (Si)
Smoothness
prior
+ τ S ∗
Low rank
prior
Our Algorithm
Initialize R and S using rigid factorisation.
Minimize energy via alternation:
Step 1: Rotation estimation.
Step 2: Shape estimation.
Efficient and highly parallelizable algorithm → GPU-friendly
13 / 17
Minimisation of E R, S
min
R
λ W − RS 2
F
Reprojection
error
+
i
TV (Si)
Smoothness
prior
+ τ S ∗
Low rank
prior
Step 1: Rotation estimation
Robust estimation by using dense data.
Solved via Levenberg-Marquardt algorithm.
Rotations are parametrised as quaternions.
13 / 17
Minimisation of E R, S
min
S
λ W − RS 2
F
Reprojection
error
+
i
TV (Si)
Smoothness
prior
+ τ S ∗
Low rank
prior
Step 2: Shape estimation
Convex sub-problem.
Optimisation via alternation between:
Per frame shape refinement: using primal dual algorithm
Enforcing low rank: using soft impute algorithm.
13 / 17
Results on real sequences
14 / 17
Quantitative Evaluation
Average RMS 3D reconstruction errors.
Sequence TB MP Ours Ours(τ = 0)
Non-smooth rotations 4.50% 5.13% 2.60% 3.32%
Smooth rotations 6.61% 5.81% 2.81% 3.89%
- TB: Akhter et al, Trajectory space: A dual representation for non-rigid structure from motion, PAMI’11.
- MP: Paladini et al, Optimal metric projections for deformable and articulated structure-from-motion,
IJCV’12.
15 / 17
Conclusions and future work
Conclusions:
First dense, template-free approach to Non-rigid Structure
from Motion.
Unified energy minimization for both rotation and shape
estimation.
Combination of low-rank and spatial regularization prior.
Using variational methods, we can do much more with
monocular sequences that one could expect
Future work:
Direct estimation from pixel intensities
Towards real-time: online formulation
Occlusion modelling
16 / 17
Thank You for Your Attention!
For more details and data
Visit: www.eecs.qmul.ac.uk/~rgarg
OR Come to our poster!
Questions?
Authors thank
Chris Russell and Sara Vicente for valuable discussions.
17 / 17

More Related Content

Similar to Ln l.agapito

On the Use of the \gamma-Re_\theta Transition Model for the Prediction of the...
On the Use of the \gamma-Re_\theta Transition Model for the Prediction of the...On the Use of the \gamma-Re_\theta Transition Model for the Prediction of the...
On the Use of the \gamma-Re_\theta Transition Model for the Prediction of the...João Baltazar
 
Foveated Ray Tracing for VR on Multiple GPUs
Foveated Ray Tracing for VR on Multiple GPUsFoveated Ray Tracing for VR on Multiple GPUs
Foveated Ray Tracing for VR on Multiple GPUsTakahiro Harada
 
Visual odometry & slam utilizing indoor structured environments
Visual odometry & slam utilizing indoor structured environmentsVisual odometry & slam utilizing indoor structured environments
Visual odometry & slam utilizing indoor structured environmentsNAVER Engineering
 
HardNet: Convolutional Network for Local Image Description
HardNet: Convolutional Network for Local Image DescriptionHardNet: Convolutional Network for Local Image Description
HardNet: Convolutional Network for Local Image DescriptionDmytro Mishkin
 
Theories and Engineering Technics of 2D-to-3D Back-Projection Problem
Theories and Engineering Technics of 2D-to-3D Back-Projection ProblemTheories and Engineering Technics of 2D-to-3D Back-Projection Problem
Theories and Engineering Technics of 2D-to-3D Back-Projection ProblemSeongcheol Baek
 
SIGGRAPH 2018 - Full Rays Ahead! From Raster to Real-Time Raytracing
SIGGRAPH 2018 - Full Rays Ahead! From Raster to Real-Time RaytracingSIGGRAPH 2018 - Full Rays Ahead! From Raster to Real-Time Raytracing
SIGGRAPH 2018 - Full Rays Ahead! From Raster to Real-Time RaytracingElectronic Arts / DICE
 
Self-dependent 3D face rotational alignment using the nose region
Self-dependent 3D face rotational alignment using the nose regionSelf-dependent 3D face rotational alignment using the nose region
Self-dependent 3D face rotational alignment using the nose regionMehryar (Mike) E., Ph.D.
 
Light Field Networks: Neural Scene Representations with Single-Evaluation Ren...
Light Field Networks: Neural Scene Representations with Single-Evaluation Ren...Light Field Networks: Neural Scene Representations with Single-Evaluation Ren...
Light Field Networks: Neural Scene Representations with Single-Evaluation Ren...Vincent Sitzmann
 
Dual-hop Variable-Gain Relaying with Beamforming over 휿−흁 Shadowed Fading Cha...
Dual-hop Variable-Gain Relaying with Beamforming over 휿−흁 Shadowed Fading Cha...Dual-hop Variable-Gain Relaying with Beamforming over 휿−흁 Shadowed Fading Cha...
Dual-hop Variable-Gain Relaying with Beamforming over 휿−흁 Shadowed Fading Cha...zeenta zeenta
 
Algorithm for the Dynamic Analysis of Plane Rectangular Rigid Frame Subjected...
Algorithm for the Dynamic Analysis of Plane Rectangular Rigid Frame Subjected...Algorithm for the Dynamic Analysis of Plane Rectangular Rigid Frame Subjected...
Algorithm for the Dynamic Analysis of Plane Rectangular Rigid Frame Subjected...Oyeniyi Samuel
 
Deep Learning Tomography
Deep Learning TomographyDeep Learning Tomography
Deep Learning TomographyAmir Adler
 
SVD and the Netflix Dataset
SVD and the Netflix DatasetSVD and the Netflix Dataset
SVD and the Netflix DatasetBen Mabey
 
Keynote - SPIE Stereoscopic Displays & Applications 2014
Keynote - SPIE Stereoscopic Displays & Applications 2014Keynote - SPIE Stereoscopic Displays & Applications 2014
Keynote - SPIE Stereoscopic Displays & Applications 2014Gordon Wetzstein
 
Design of Flexible Pavement Using AASHTO.pptx
Design of Flexible Pavement Using AASHTO.pptxDesign of Flexible Pavement Using AASHTO.pptx
Design of Flexible Pavement Using AASHTO.pptxmohammeed3
 
Model-counting Approaches For Nonlinear Numerical Constraints
Model-counting Approaches For Nonlinear Numerical ConstraintsModel-counting Approaches For Nonlinear Numerical Constraints
Model-counting Approaches For Nonlinear Numerical ConstraintsQuoc-Sang Phan
 
Super resolution in deep learning era - Jaejun Yoo
Super resolution in deep learning era - Jaejun YooSuper resolution in deep learning era - Jaejun Yoo
Super resolution in deep learning era - Jaejun YooJaeJun Yoo
 
Near Surface Geoscience Conference 2015, Turin - A Spatial Velocity Analysis ...
Near Surface Geoscience Conference 2015, Turin - A Spatial Velocity Analysis ...Near Surface Geoscience Conference 2015, Turin - A Spatial Velocity Analysis ...
Near Surface Geoscience Conference 2015, Turin - A Spatial Velocity Analysis ...CRS4 Research Center in Sardinia
 
Fixed Point Realization of Iterative LR-Aided Soft MIMO Decoding Algorithm
Fixed Point Realization of Iterative LR-Aided Soft MIMO Decoding AlgorithmFixed Point Realization of Iterative LR-Aided Soft MIMO Decoding Algorithm
Fixed Point Realization of Iterative LR-Aided Soft MIMO Decoding AlgorithmCSCJournals
 

Similar to Ln l.agapito (20)

On the Use of the \gamma-Re_\theta Transition Model for the Prediction of the...
On the Use of the \gamma-Re_\theta Transition Model for the Prediction of the...On the Use of the \gamma-Re_\theta Transition Model for the Prediction of the...
On the Use of the \gamma-Re_\theta Transition Model for the Prediction of the...
 
Foveated Ray Tracing for VR on Multiple GPUs
Foveated Ray Tracing for VR on Multiple GPUsFoveated Ray Tracing for VR on Multiple GPUs
Foveated Ray Tracing for VR on Multiple GPUs
 
ICRA Nathan Piasco
ICRA Nathan PiascoICRA Nathan Piasco
ICRA Nathan Piasco
 
Visual odometry & slam utilizing indoor structured environments
Visual odometry & slam utilizing indoor structured environmentsVisual odometry & slam utilizing indoor structured environments
Visual odometry & slam utilizing indoor structured environments
 
HardNet: Convolutional Network for Local Image Description
HardNet: Convolutional Network for Local Image DescriptionHardNet: Convolutional Network for Local Image Description
HardNet: Convolutional Network for Local Image Description
 
Theories and Engineering Technics of 2D-to-3D Back-Projection Problem
Theories and Engineering Technics of 2D-to-3D Back-Projection ProblemTheories and Engineering Technics of 2D-to-3D Back-Projection Problem
Theories and Engineering Technics of 2D-to-3D Back-Projection Problem
 
SIGGRAPH 2018 - Full Rays Ahead! From Raster to Real-Time Raytracing
SIGGRAPH 2018 - Full Rays Ahead! From Raster to Real-Time RaytracingSIGGRAPH 2018 - Full Rays Ahead! From Raster to Real-Time Raytracing
SIGGRAPH 2018 - Full Rays Ahead! From Raster to Real-Time Raytracing
 
Self-dependent 3D face rotational alignment using the nose region
Self-dependent 3D face rotational alignment using the nose regionSelf-dependent 3D face rotational alignment using the nose region
Self-dependent 3D face rotational alignment using the nose region
 
Light Field Networks: Neural Scene Representations with Single-Evaluation Ren...
Light Field Networks: Neural Scene Representations with Single-Evaluation Ren...Light Field Networks: Neural Scene Representations with Single-Evaluation Ren...
Light Field Networks: Neural Scene Representations with Single-Evaluation Ren...
 
Human Action Recognition
Human Action RecognitionHuman Action Recognition
Human Action Recognition
 
Dual-hop Variable-Gain Relaying with Beamforming over 휿−흁 Shadowed Fading Cha...
Dual-hop Variable-Gain Relaying with Beamforming over 휿−흁 Shadowed Fading Cha...Dual-hop Variable-Gain Relaying with Beamforming over 휿−흁 Shadowed Fading Cha...
Dual-hop Variable-Gain Relaying with Beamforming over 휿−흁 Shadowed Fading Cha...
 
Algorithm for the Dynamic Analysis of Plane Rectangular Rigid Frame Subjected...
Algorithm for the Dynamic Analysis of Plane Rectangular Rigid Frame Subjected...Algorithm for the Dynamic Analysis of Plane Rectangular Rigid Frame Subjected...
Algorithm for the Dynamic Analysis of Plane Rectangular Rigid Frame Subjected...
 
Deep Learning Tomography
Deep Learning TomographyDeep Learning Tomography
Deep Learning Tomography
 
SVD and the Netflix Dataset
SVD and the Netflix DatasetSVD and the Netflix Dataset
SVD and the Netflix Dataset
 
Keynote - SPIE Stereoscopic Displays & Applications 2014
Keynote - SPIE Stereoscopic Displays & Applications 2014Keynote - SPIE Stereoscopic Displays & Applications 2014
Keynote - SPIE Stereoscopic Displays & Applications 2014
 
Design of Flexible Pavement Using AASHTO.pptx
Design of Flexible Pavement Using AASHTO.pptxDesign of Flexible Pavement Using AASHTO.pptx
Design of Flexible Pavement Using AASHTO.pptx
 
Model-counting Approaches For Nonlinear Numerical Constraints
Model-counting Approaches For Nonlinear Numerical ConstraintsModel-counting Approaches For Nonlinear Numerical Constraints
Model-counting Approaches For Nonlinear Numerical Constraints
 
Super resolution in deep learning era - Jaejun Yoo
Super resolution in deep learning era - Jaejun YooSuper resolution in deep learning era - Jaejun Yoo
Super resolution in deep learning era - Jaejun Yoo
 
Near Surface Geoscience Conference 2015, Turin - A Spatial Velocity Analysis ...
Near Surface Geoscience Conference 2015, Turin - A Spatial Velocity Analysis ...Near Surface Geoscience Conference 2015, Turin - A Spatial Velocity Analysis ...
Near Surface Geoscience Conference 2015, Turin - A Spatial Velocity Analysis ...
 
Fixed Point Realization of Iterative LR-Aided Soft MIMO Decoding Algorithm
Fixed Point Realization of Iterative LR-Aided Soft MIMO Decoding AlgorithmFixed Point Realization of Iterative LR-Aided Soft MIMO Decoding Algorithm
Fixed Point Realization of Iterative LR-Aided Soft MIMO Decoding Algorithm
 

More from potaters

Image formation
Image formationImage formation
Image formationpotaters
 
Motion and tracking
Motion and trackingMotion and tracking
Motion and trackingpotaters
 
BMVA summer school MATLAB programming tutorial
BMVA summer school MATLAB programming tutorialBMVA summer school MATLAB programming tutorial
BMVA summer school MATLAB programming tutorialpotaters
 
Statistical models of shape and appearance
Statistical models of shape and appearanceStatistical models of shape and appearance
Statistical models of shape and appearancepotaters
 
Vision Algorithmics
Vision AlgorithmicsVision Algorithmics
Vision Algorithmicspotaters
 
Performance characterization in computer vision
Performance characterization in computer visionPerformance characterization in computer vision
Performance characterization in computer visionpotaters
 
Machine learning for computer vision part 2
Machine learning for computer vision part 2Machine learning for computer vision part 2
Machine learning for computer vision part 2potaters
 
Machine learning fro computer vision - a whirlwind of key concepts for the un...
Machine learning fro computer vision - a whirlwind of key concepts for the un...Machine learning fro computer vision - a whirlwind of key concepts for the un...
Machine learning fro computer vision - a whirlwind of key concepts for the un...potaters
 
Low level vision - A tuturial
Low level vision - A tuturialLow level vision - A tuturial
Low level vision - A tuturialpotaters
 
Local feature descriptors for visual recognition
Local feature descriptors for visual recognitionLocal feature descriptors for visual recognition
Local feature descriptors for visual recognitionpotaters
 
Image segmentation
Image segmentationImage segmentation
Image segmentationpotaters
 
A primer for colour computer vision
A primer for colour computer visionA primer for colour computer vision
A primer for colour computer visionpotaters
 
Cognitive Vision - After the hype
Cognitive Vision - After the hypeCognitive Vision - After the hype
Cognitive Vision - After the hypepotaters
 
Graphical Models for chains, trees and grids
Graphical Models for chains, trees and gridsGraphical Models for chains, trees and grids
Graphical Models for chains, trees and gridspotaters
 
Medical image computing - BMVA summer school 2014
Medical image computing - BMVA summer school 2014Medical image computing - BMVA summer school 2014
Medical image computing - BMVA summer school 2014potaters
 
Decision Forests and discriminant analysis
Decision Forests and discriminant analysisDecision Forests and discriminant analysis
Decision Forests and discriminant analysispotaters
 

More from potaters (16)

Image formation
Image formationImage formation
Image formation
 
Motion and tracking
Motion and trackingMotion and tracking
Motion and tracking
 
BMVA summer school MATLAB programming tutorial
BMVA summer school MATLAB programming tutorialBMVA summer school MATLAB programming tutorial
BMVA summer school MATLAB programming tutorial
 
Statistical models of shape and appearance
Statistical models of shape and appearanceStatistical models of shape and appearance
Statistical models of shape and appearance
 
Vision Algorithmics
Vision AlgorithmicsVision Algorithmics
Vision Algorithmics
 
Performance characterization in computer vision
Performance characterization in computer visionPerformance characterization in computer vision
Performance characterization in computer vision
 
Machine learning for computer vision part 2
Machine learning for computer vision part 2Machine learning for computer vision part 2
Machine learning for computer vision part 2
 
Machine learning fro computer vision - a whirlwind of key concepts for the un...
Machine learning fro computer vision - a whirlwind of key concepts for the un...Machine learning fro computer vision - a whirlwind of key concepts for the un...
Machine learning fro computer vision - a whirlwind of key concepts for the un...
 
Low level vision - A tuturial
Low level vision - A tuturialLow level vision - A tuturial
Low level vision - A tuturial
 
Local feature descriptors for visual recognition
Local feature descriptors for visual recognitionLocal feature descriptors for visual recognition
Local feature descriptors for visual recognition
 
Image segmentation
Image segmentationImage segmentation
Image segmentation
 
A primer for colour computer vision
A primer for colour computer visionA primer for colour computer vision
A primer for colour computer vision
 
Cognitive Vision - After the hype
Cognitive Vision - After the hypeCognitive Vision - After the hype
Cognitive Vision - After the hype
 
Graphical Models for chains, trees and grids
Graphical Models for chains, trees and gridsGraphical Models for chains, trees and grids
Graphical Models for chains, trees and grids
 
Medical image computing - BMVA summer school 2014
Medical image computing - BMVA summer school 2014Medical image computing - BMVA summer school 2014
Medical image computing - BMVA summer school 2014
 
Decision Forests and discriminant analysis
Decision Forests and discriminant analysisDecision Forests and discriminant analysis
Decision Forests and discriminant analysis
 

Recently uploaded

User Guide: Magellan MX™ Weather Station
User Guide: Magellan MX™ Weather StationUser Guide: Magellan MX™ Weather Station
User Guide: Magellan MX™ Weather StationColumbia Weather Systems
 
User Guide: Orion™ Weather Station (Columbia Weather Systems)
User Guide: Orion™ Weather Station (Columbia Weather Systems)User Guide: Orion™ Weather Station (Columbia Weather Systems)
User Guide: Orion™ Weather Station (Columbia Weather Systems)Columbia Weather Systems
 
Manassas R - Parkside Middle School 🌎🏫
Manassas R - Parkside Middle School 🌎🏫Manassas R - Parkside Middle School 🌎🏫
Manassas R - Parkside Middle School 🌎🏫qfactory1
 
GLYCOSIDES Classification Of GLYCOSIDES Chemical Tests Glycosides
GLYCOSIDES Classification Of GLYCOSIDES  Chemical Tests GlycosidesGLYCOSIDES Classification Of GLYCOSIDES  Chemical Tests Glycosides
GLYCOSIDES Classification Of GLYCOSIDES Chemical Tests GlycosidesNandakishor Bhaurao Deshmukh
 
The dark energy paradox leads to a new structure of spacetime.pptx
The dark energy paradox leads to a new structure of spacetime.pptxThe dark energy paradox leads to a new structure of spacetime.pptx
The dark energy paradox leads to a new structure of spacetime.pptxEran Akiva Sinbar
 
Pests of soyabean_Binomics_IdentificationDr.UPR.pdf
Pests of soyabean_Binomics_IdentificationDr.UPR.pdfPests of soyabean_Binomics_IdentificationDr.UPR.pdf
Pests of soyabean_Binomics_IdentificationDr.UPR.pdfPirithiRaju
 
Citronella presentation SlideShare mani upadhyay
Citronella presentation SlideShare mani upadhyayCitronella presentation SlideShare mani upadhyay
Citronella presentation SlideShare mani upadhyayupadhyaymani499
 
Pests of castor_Binomics_Identification_Dr.UPR.pdf
Pests of castor_Binomics_Identification_Dr.UPR.pdfPests of castor_Binomics_Identification_Dr.UPR.pdf
Pests of castor_Binomics_Identification_Dr.UPR.pdfPirithiRaju
 
Pests of Bengal gram_Identification_Dr.UPR.pdf
Pests of Bengal gram_Identification_Dr.UPR.pdfPests of Bengal gram_Identification_Dr.UPR.pdf
Pests of Bengal gram_Identification_Dr.UPR.pdfPirithiRaju
 
Pests of jatropha_Bionomics_identification_Dr.UPR.pdf
Pests of jatropha_Bionomics_identification_Dr.UPR.pdfPests of jatropha_Bionomics_identification_Dr.UPR.pdf
Pests of jatropha_Bionomics_identification_Dr.UPR.pdfPirithiRaju
 
Microteaching on terms used in filtration .Pharmaceutical Engineering
Microteaching on terms used in filtration .Pharmaceutical EngineeringMicroteaching on terms used in filtration .Pharmaceutical Engineering
Microteaching on terms used in filtration .Pharmaceutical EngineeringPrajakta Shinde
 
User Guide: Capricorn FLX™ Weather Station
User Guide: Capricorn FLX™ Weather StationUser Guide: Capricorn FLX™ Weather Station
User Guide: Capricorn FLX™ Weather StationColumbia Weather Systems
 
User Guide: Pulsar™ Weather Station (Columbia Weather Systems)
User Guide: Pulsar™ Weather Station (Columbia Weather Systems)User Guide: Pulsar™ Weather Station (Columbia Weather Systems)
User Guide: Pulsar™ Weather Station (Columbia Weather Systems)Columbia Weather Systems
 
Servosystem Theory / Cybernetic Theory by Petrovic
Servosystem Theory / Cybernetic Theory by PetrovicServosystem Theory / Cybernetic Theory by Petrovic
Servosystem Theory / Cybernetic Theory by PetrovicAditi Jain
 
PROJECTILE MOTION-Horizontal and Vertical
PROJECTILE MOTION-Horizontal and VerticalPROJECTILE MOTION-Horizontal and Vertical
PROJECTILE MOTION-Horizontal and VerticalMAESTRELLAMesa2
 
Speech, hearing, noise, intelligibility.pptx
Speech, hearing, noise, intelligibility.pptxSpeech, hearing, noise, intelligibility.pptx
Speech, hearing, noise, intelligibility.pptxpriyankatabhane
 
trihybrid cross , test cross chi squares
trihybrid cross , test cross chi squarestrihybrid cross , test cross chi squares
trihybrid cross , test cross chi squaresusmanzain586
 
Observational constraints on mergers creating magnetism in massive stars
Observational constraints on mergers creating magnetism in massive starsObservational constraints on mergers creating magnetism in massive stars
Observational constraints on mergers creating magnetism in massive starsSérgio Sacani
 

Recently uploaded (20)

User Guide: Magellan MX™ Weather Station
User Guide: Magellan MX™ Weather StationUser Guide: Magellan MX™ Weather Station
User Guide: Magellan MX™ Weather Station
 
User Guide: Orion™ Weather Station (Columbia Weather Systems)
User Guide: Orion™ Weather Station (Columbia Weather Systems)User Guide: Orion™ Weather Station (Columbia Weather Systems)
User Guide: Orion™ Weather Station (Columbia Weather Systems)
 
Manassas R - Parkside Middle School 🌎🏫
Manassas R - Parkside Middle School 🌎🏫Manassas R - Parkside Middle School 🌎🏫
Manassas R - Parkside Middle School 🌎🏫
 
GLYCOSIDES Classification Of GLYCOSIDES Chemical Tests Glycosides
GLYCOSIDES Classification Of GLYCOSIDES  Chemical Tests GlycosidesGLYCOSIDES Classification Of GLYCOSIDES  Chemical Tests Glycosides
GLYCOSIDES Classification Of GLYCOSIDES Chemical Tests Glycosides
 
The dark energy paradox leads to a new structure of spacetime.pptx
The dark energy paradox leads to a new structure of spacetime.pptxThe dark energy paradox leads to a new structure of spacetime.pptx
The dark energy paradox leads to a new structure of spacetime.pptx
 
Pests of soyabean_Binomics_IdentificationDr.UPR.pdf
Pests of soyabean_Binomics_IdentificationDr.UPR.pdfPests of soyabean_Binomics_IdentificationDr.UPR.pdf
Pests of soyabean_Binomics_IdentificationDr.UPR.pdf
 
Citronella presentation SlideShare mani upadhyay
Citronella presentation SlideShare mani upadhyayCitronella presentation SlideShare mani upadhyay
Citronella presentation SlideShare mani upadhyay
 
Let’s Say Someone Did Drop the Bomb. Then What?
Let’s Say Someone Did Drop the Bomb. Then What?Let’s Say Someone Did Drop the Bomb. Then What?
Let’s Say Someone Did Drop the Bomb. Then What?
 
Pests of castor_Binomics_Identification_Dr.UPR.pdf
Pests of castor_Binomics_Identification_Dr.UPR.pdfPests of castor_Binomics_Identification_Dr.UPR.pdf
Pests of castor_Binomics_Identification_Dr.UPR.pdf
 
Pests of Bengal gram_Identification_Dr.UPR.pdf
Pests of Bengal gram_Identification_Dr.UPR.pdfPests of Bengal gram_Identification_Dr.UPR.pdf
Pests of Bengal gram_Identification_Dr.UPR.pdf
 
Pests of jatropha_Bionomics_identification_Dr.UPR.pdf
Pests of jatropha_Bionomics_identification_Dr.UPR.pdfPests of jatropha_Bionomics_identification_Dr.UPR.pdf
Pests of jatropha_Bionomics_identification_Dr.UPR.pdf
 
AZOTOBACTER AS BIOFERILIZER.PPTX
AZOTOBACTER AS BIOFERILIZER.PPTXAZOTOBACTER AS BIOFERILIZER.PPTX
AZOTOBACTER AS BIOFERILIZER.PPTX
 
Microteaching on terms used in filtration .Pharmaceutical Engineering
Microteaching on terms used in filtration .Pharmaceutical EngineeringMicroteaching on terms used in filtration .Pharmaceutical Engineering
Microteaching on terms used in filtration .Pharmaceutical Engineering
 
User Guide: Capricorn FLX™ Weather Station
User Guide: Capricorn FLX™ Weather StationUser Guide: Capricorn FLX™ Weather Station
User Guide: Capricorn FLX™ Weather Station
 
User Guide: Pulsar™ Weather Station (Columbia Weather Systems)
User Guide: Pulsar™ Weather Station (Columbia Weather Systems)User Guide: Pulsar™ Weather Station (Columbia Weather Systems)
User Guide: Pulsar™ Weather Station (Columbia Weather Systems)
 
Servosystem Theory / Cybernetic Theory by Petrovic
Servosystem Theory / Cybernetic Theory by PetrovicServosystem Theory / Cybernetic Theory by Petrovic
Servosystem Theory / Cybernetic Theory by Petrovic
 
PROJECTILE MOTION-Horizontal and Vertical
PROJECTILE MOTION-Horizontal and VerticalPROJECTILE MOTION-Horizontal and Vertical
PROJECTILE MOTION-Horizontal and Vertical
 
Speech, hearing, noise, intelligibility.pptx
Speech, hearing, noise, intelligibility.pptxSpeech, hearing, noise, intelligibility.pptx
Speech, hearing, noise, intelligibility.pptx
 
trihybrid cross , test cross chi squares
trihybrid cross , test cross chi squarestrihybrid cross , test cross chi squares
trihybrid cross , test cross chi squares
 
Observational constraints on mergers creating magnetism in massive stars
Observational constraints on mergers creating magnetism in massive starsObservational constraints on mergers creating magnetism in massive stars
Observational constraints on mergers creating magnetism in massive stars
 

Ln l.agapito

  • 1. Dense Variational Reconstruction of Non-Rigid Surfaces from Monocular Video Ravi Garg Anastasios Roussos∗ Lourdes Agapito∗ Queen Mary, University of London ∗ Now at UCL Before This Paper 1 / 17
  • 2. Dense Non Rigid Structure from Motion Input: monocular sequence of non-rigid surface. ... Goal: dense 3D reconstruction for every frame. 2 / 17
  • 3. Dense Non Rigid Structure from Motion Input: monocular sequence of non-rigid surface. ... Goal: dense 3D reconstruction for every frame. 2 / 17
  • 4. Dense Non Rigid Structure from Motion Input: monocular sequence of non-rigid surface. ... Goal: dense 3D reconstruction for every frame. 2 / 17
  • 5. Dense Non Rigid Structure from Motion Input: monocular sequence of non-rigid surface. ... Goal: dense 3D reconstruction for every frame. 2 / 17
  • 6. Dense Non Rigid Structure from Motion Input: monocular sequence of non-rigid surface. ... Goal: dense 3D reconstruction for every frame. 2 / 17
  • 7. Dense Non Rigid Structure from Motion Input: monocular sequence of non-rigid surface. ... Goal: dense 3D reconstruction for every frame. Goal: dense 3D reconstruction for every frame. NO additional sensors. NO pre-trained shape models. NO surface template. 2 / 17
  • 8. Dense Non Rigid Structure from Motion Input: monocular sequence of non-rigid surface. ... NRSfM: ill posed problem Goal: dense 3D reconstruction for every frame. Goal: dense 3D reconstruction for every frame. NO additional sensors. NO pre-trained shape models. NO surface template. 2 / 17
  • 9. Traditional Sparse Non Rigid Structure from Motion ... 3 / 17
  • 10. Traditional Sparse Non Rigid Structure from Motion ...Feature Tracking ... 3 / 17
  • 11. Traditional Sparse Non Rigid Structure from Motion ...Feature Tracking ... ...3D Shape Inference 3 / 17
  • 12. Traditional Sparse Non Rigid Structure from Motion Priors ...Feature Tracking ... + ...3D Shape Inference 3 / 17
  • 13. Low Rank Prior for NRSfM Shape Space (Bregler, Hertzmann, Biermann, Recovering non-rigid 3D shape from image streams CVPR’00.) Bregler et al. CVPR’00, Brand CVPR’01, Xiao et al. IJCV’06, Torresani et al. PAMI’08, Akhter et al. CVPR’09, Bartoli et al. CVPR2008, Paladini et al. IJCV’12,Dai et al. CVPR’12... 4 / 17
  • 14. Low Rank Prior for NRSfM Shape Space (Bregler, Hertzmann, Biermann, Recovering non-rigid 3D shape from image streams CVPR’00.) (Park et al. ECCV’10) 4 / 17
  • 15. Inspiration from Dense Rigid Reconstruction (Newcombe, Lovegrove, Davison, DTAM: Dense Tracking and Mapping in Real-Time, ICCV’11) 5 / 17
  • 16. Inspiration from Dense Rigid Reconstruction (Newcombe, Lovegrove, Davison, DTAM: Dense Tracking and Mapping in Real-Time, ICCV’11) Key features Variational approach. Use of smoothness priors. Per pixel reconstruction. Scalable and GPU friendly algorithm. 5 / 17
  • 17. Leap from sparse to dense NRSfM Sparse Dai et al. CVPR’12 Dense This work 6 / 17
  • 18. Leap from sparse to dense NRSfM Sparse Dai et al. CVPR’12 Dense This work We take the best of both worlds: Low rank prior from sparse non rigid SfM. Variational framework from dense rigid SfM. 6 / 17
  • 19. Leap from sparse to dense NRSfM Sparse Dai et al. CVPR’12 Dense This work We take the best of both worlds: Low rank prior from sparse non rigid SfM. Variational framework from dense rigid SfM. Our contribution First variational formulation to dense NRSfM. Scalable algorithm which can be ported on GPU. 6 / 17
  • 20. Our Approach in a Nutshell ... 7 / 17
  • 21. Our Approach in a Nutshell ... ... Step 1: Dense Video Registration ... 7 / 17
  • 22. Our Approach in a Nutshell Step 2: Dense Shape Inference ... ... ... Step 1: Dense Video Registration ... 7 / 17
  • 23. Our Approach in a Nutshell Step 2: Dense Shape Inference ... ... ... Step 1: Dense Video Registration ... Priors+ 7 / 17
  • 24. Our Approach in a Nutshell Step 2: Dense Shape Inference ... ... ... Step 1: Dense Video Registration ... Low rank. Spatial smoothness.+ 7 / 17
  • 25. Our Approach in a Nutshell ... ... Low rank. Spatial smoothness. Step 1: Dense Video Registration ... + Garg, Roussos, Agapito, A variational approach to video registration with subspace constraints, IJCV’13. 7 / 17
  • 26. Our Approach in a Nutshell Step 2: Dense Shape Inference ... ... Low rank. Spatial smoothness.+ 7 / 17
  • 32. Energy Minimisation Approach to NRSfM Formulation of a single unified energy to estimate: Orthographic projection matrices 3D shapes for all the frames E R , S = λ Edata R, S + Ereg S + τ Etrace S reprojection error over all frames spatial smoothness prior on 3D shapes low rank prior on 3D shapes 9 / 17
  • 33. Energy Minimisation Approach to NRSfM Formulation of a single unified energy to estimate: Orthographic projection matrices 3D shapes for all the frames E R , S = λ Edata R, S + Ereg S + τ Etrace S reprojection error over all frames spatial smoothness prior on 3D shapes low rank prior on 3D shapes 9 / 17
  • 34. Energy Minimisation Approach to NRSfM Formulation of a single unified energy to estimate: Orthographic projection matrices 3D shapes for all the frames E R , S = λ Edata R, S + Ereg S + τ Etrace S reprojection error over all frames spatial smoothness prior on 3D shapes low rank prior on 3D shapes 9 / 17
  • 35. Energy Minimisation Approach to NRSfM Formulation of a single unified energy to estimate: Orthographic projection matrices 3D shapes for all the frames E R , S = λ Edata R, S + Ereg S + τ Etrace S reprojection error over all frames spatial smoothness prior on 3D shapes low rank prior on 3D shapes 9 / 17
  • 36. Reprojection Error E ` R, S ´ = λEdata ` R, S ´ + Ereg ` S ´ + τEtrace ` S ´ Edata (R, S) = W − RS 2 F 10 / 17
  • 37. Spatial Smoothness Prior E ` R, S ´ = λEdata ` R, S ´ + Ereg ` S ´ + τEtrace ` S ´ Ereg S = i TV (Si) −−−−−−→ Without regularisation With regularisation 11 / 17
  • 38. Low Rank Prior E ` R, S ´ = λEdata ` R, S ´ + Ereg ` S ´ + τEtrace ` S ´ Etrace S = S ∗ = i σi(S) lies in−−−−→ span K F Angst et al. ECCV’12, Dai et al. CVPR’12, Angst et al. ICCV’11, Dai et al. ECCV’10 12 / 17
  • 39. Minimisation of E R, S min R,S λ W − RS 2 F Reprojection error + i TV (Si) Smoothness prior + τ S ∗ Low rank prior 13 / 17
  • 40. Minimisation of E R, S min R,S λ W − RS 2 F Reprojection error + i TV (Si) Smoothness prior + τ S ∗ Low rank prior Our Algorithm Initialize R and S using rigid factorisation. Minimize energy via alternation: Step 1: Rotation estimation. Step 2: Shape estimation. Efficient and highly parallelizable algorithm → GPU-friendly 13 / 17
  • 41. Minimisation of E R, S min R λ W − RS 2 F Reprojection error + i TV (Si) Smoothness prior + τ S ∗ Low rank prior Step 1: Rotation estimation Robust estimation by using dense data. Solved via Levenberg-Marquardt algorithm. Rotations are parametrised as quaternions. 13 / 17
  • 42. Minimisation of E R, S min S λ W − RS 2 F Reprojection error + i TV (Si) Smoothness prior + τ S ∗ Low rank prior Step 2: Shape estimation Convex sub-problem. Optimisation via alternation between: Per frame shape refinement: using primal dual algorithm Enforcing low rank: using soft impute algorithm. 13 / 17
  • 43. Results on real sequences 14 / 17
  • 44. Quantitative Evaluation Average RMS 3D reconstruction errors. Sequence TB MP Ours Ours(τ = 0) Non-smooth rotations 4.50% 5.13% 2.60% 3.32% Smooth rotations 6.61% 5.81% 2.81% 3.89% - TB: Akhter et al, Trajectory space: A dual representation for non-rigid structure from motion, PAMI’11. - MP: Paladini et al, Optimal metric projections for deformable and articulated structure-from-motion, IJCV’12. 15 / 17
  • 45. Conclusions and future work Conclusions: First dense, template-free approach to Non-rigid Structure from Motion. Unified energy minimization for both rotation and shape estimation. Combination of low-rank and spatial regularization prior. Using variational methods, we can do much more with monocular sequences that one could expect Future work: Direct estimation from pixel intensities Towards real-time: online formulation Occlusion modelling 16 / 17
  • 46. Thank You for Your Attention! For more details and data Visit: www.eecs.qmul.ac.uk/~rgarg OR Come to our poster! Questions? Authors thank Chris Russell and Sara Vicente for valuable discussions. 17 / 17