single-view and multi-view planar models for dense ...csadc/lpm17/slides/lpm2017slides-civera… ·...

28
Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha, José M. Fácil and Javier Civera SLAMLab – Robotics, Perception and Real-Time Group Universidad de Zaragoza, Spain International Workshop on Lines, Planes and Manhattan Models for 3-D Mapping (LPM 2017) September 28, 2017, IROS 2017, Vancouver.

Upload: others

Post on 10-Jun-2020

4 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Single-View and Multi-View Planar Models for Dense Monocular Mapping

Alejo Concha, José M. Fácil and Javier Civera

SLAMLab – Robotics, Perception and Real-Time Group Universidad de Zaragoza, Spain

International Workshop on Lines, Planes and Manhattan Models

for 3-D Mapping (LPM 2017) September 28, 2017, IROS 2017, Vancouver.

Page 2: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Index

• Motivation • Background (direct mapping)

• Dense monocular mapping.

• Superpixels in monocular mapping • Superpixel triangulation. • Dense mapping using superpixels. • Superpixel fitting.

• Learning-based planar models in monocular mapping

• Data-driven primitives • Layout • Deep models

• Conclusions

Page 3: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Motivation

• The scene model is limited in feature-based monocular SLAM.

• Our goal: Dense mapping from monocular (RGB) image sequences

Page 4: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Background: Dense Monocular Mapping High Texture Low Texture

Accuracy Density Cost Accuracy Density Cost

Sparse/Semi-d.

Dense

Page 5: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Dense Monocular Mapping: Low Texture

High Texture Low Texture

Accuracy Density Cost Accuracy Density Cost

Dense

Page 6: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Superpixels (mid-level)

High Texture Low Texture

Accuracy Density Cost Accuracy Density Cost

Sparse/Semi-d.

Dense

Superpixels

Dense + Sup.

• Image segmentation based on color and 2D distance.

• Decent features for textureless areas • We assume that homogeneous color

regions are almost planar.

Page 7: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

High Texture Low Texture

Accuracy Density Cost Accuracy Density Cost

Dense

Dense Mapping: Low Texture

Page 8: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Semi-dense Mapping: Low Texture

High Texture Low Texture

Accuracy Density Cost Accuracy Density Cost

Sparse/Semi-d.

Page 9: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

2D Superpixels: Low Texture

High Texture Low Texture

Accuracy Density Cost Accuracy Density Cost

Superpixels

Page 10: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Superpixel Triangulation

H

Multiview model: Homography (ℎ = 𝐾(𝑅 + 𝑡𝑛/𝑑)𝐾−1)

Error: Contour reprojection error (ɛ)

Montecarlo Initialization: For every superpixel we create several reasonable {𝑛, 𝑑} hypothesis and rank them by their error.

Page 11: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Superpixel Triangulation

Multiview model: Homography (ℎ = 𝐾(𝑅 + 𝑡𝑛/𝑑)𝐾−1)

Error: Contour reprojection error (ɛ)

Mapping: Minimize the reprojection error.

H

Page 12: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Superpixels in low-textured areas

High Texture Low Texture

Accuracy Density Cost Accuracy Density Cost

Superpixels

Page 13: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Using Superpixels in Monocular SLAM

Page 14: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Dense + Superpixels

Page 15: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Dense + Superpixels

High Texture Low Texture

Accuracy Density Cost Accuracy Density Cost

Dense + Sup.

Page 16: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

PMVS (high-gradient pixels) Dense (TV-regularization)

Superpixels PMVS + Superpixels Dense + Superpixels

Video (input)

Dense + Superpixels (5 centimetres error!)

Alejo Concha and Javier Civera. Using Superpixels in Monocular SLAM. ICRA 2014

Yasutaka Furukawa and Jean Ponce. Accurate, dense, and robust multiview stereopsis. IEEE Transactions on Pattern Analysis and Machine Intelligence, 32(8):13621376, 2010.

Based on Richard A Newcombe, Steven J Lovegrove, and Andrew J Davison. Dtam: Dense tracking and mapping in real-time. In Computer Vision (ICCV), 2011 IEEE International Conference on, pages 23202327. IEEE, 2011.

Alejo Concha, Wajahat Hussain, Luis Montano and Javier Civera, Manhattan and Piecewise-Planar Constraints for Dense Monocular Mapping, RSS 2014.

Page 17: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Fitting 3D Superpixels to Semi-dense Maps

• TV-regularization is expensive, GPU might be needed for real-time. • Semidense mapping and superpixels is a reasonable option cheaper than

TV-regularization (CPU) and with a small loss on density. • Having a semidense map superpixels can be initialized via SVD more

accurately and at a lower cost. • LIMITATION: We need parallax!!

Code at https://github.com/alejocb/dpptam

Page 18: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Data-driven primitives (mid-level)

.

Feature discovery on RGB-D training data.

Extracts patterns that are consistent in D and discriminative in RGB

At test time, from a single RGB view we can predict mid-level depth patterns.

Page 19: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Multiview Layout (high-level) (a) Sparse/Semidense reconstruction. (b) Plane normals from 3D vanishing points (image VP, backprojection, 3D clustering). (c) Plane distances from a sparse/semidense multiview reconstruction. (d) Superpixel segmentation, geometric and photometric feature extraction. (e), (f) Classification (Adaboost)

Page 20: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Superpixels, Data-Driven Primitives and Layout

Page 21: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Superpixels, Data-Driven Primitives and Layout

• NYU dataset, high-parallax sequences

Page 22: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Superpixels, Data-Driven Primitives and Layout

• NYU dataset, low-parallax sequences

Page 23: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Single-View Depth Prediction • Several networks already

exist (Eigen14, Eigen15, Liu15, Liu15, Chakrabarti16, Cao16, Godard16, Ummenhofer 16…)

Page 24: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Deep Learning Depth vs. Multiview Depth

Deep Learning Depth Multiview Depth

Fairly accurate in all pixels Very accurate in high-gradient pixels,

inaccurate in low-gradient ones

Fairly accurate for single view Very accurate for high-parallax motion,

inaccurate for low-parallax one

No model for the error Good model for the error

Approximate scale 3D reconstruction up to scale

Errors depend on the image content Errors depend on the geometry

Page 25: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Fusing depth from deep learning and multiple views

• The fusion is not trivial. • No uncertainty for CNN

depth. • Errors come from different

sources.

• Our assumption is • In general, deep learning depth

is more accurate • Multiple view more accurate

for high texture - high parallax

Page 26: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Results

• The error of deep learning depth is ~50% lower than multi-view one. • Our fusion reduces the error ~10% over the deep learning results. • The scale invariant metric shows that our fusion fixes the structure. • Deep depth generalizes well (Eigen15 was trained on NYU but is accurate

on TUM)

Page 27: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,

Conclusions (no free lunch!)

Point-based features (low-level)

High accuracy iff ↑texture and ↑parallax.

Superpixels (mid-level)

High accuracy iff ↓texture and ↑parallax.

Data-driven primitives (mid-level)

Fair accuracy for → ↑ texture and ↓parallax.

Not fully dense.

Layout (high-level)

Fair accuracy even for ↓texture and ↓parallax.

Assumes a predetermined scene shape.

Deep learning (mid/high-level)

Fair accuracy even for ↓texture and ↓parallax.

Fully dense.

More general.

Page 28: Single-View and Multi-View Planar Models for Dense ...csadc/LPM17/slides/LPM2017slides-civera… · Single-View and Multi-View Planar Models for Dense Monocular Mapping Alejo Concha,