Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 1/33
Chapter 8
Direct Approaches to Visual SLAM
Multiple View Geometry
Summer 2021
Prof. Daniel Cremers
Chair for Computer Vision and Artificial Intelligence
Departments of Informatics & Mathematics
Technical University of Munich
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 2/33
Overview
1 Direct Methods
2 Realtime Dense Geometry
3 Dense RGB-D Tracking
4 Loop Closure and Global Consistency
5 Dense Tracking and Mapping
6 Large Scale Direct Monocular SLAM
7 Direct Sparse Odometry
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 3/33
Classical Approaches to Multiple View Reconstruction
In the past chapters we have studied classical approaches to
multiple view reconstruction. These methods tackle the
problem of structure and motion estimation (or visual SLAM) in
several steps:
1 A set of feature points is extracted from the images –
ideally points such as corners which can be reliably
identified in subsequent images as well.
2 One determines a correspondence of these points across
the various images. This can be done either through local
tracking (using optical flow approaches) or by random
sampling of possible partners based on a feature
descriptor (SIFT, SURF, etc.) associated with each point.
3 The camera motion is estimated based on a set of
corresponding points. In many approaches this is done by
a series of algorithms such as the eight-point algorithm or
the five-point algorithm followed by bundle adjustment.
4 For a given camera motion one can then compute a dense
reconstruction using stereo reconstruction methods.
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 4/33
Shortcomings of Classical Approaches
Such classical approaches are indirect in the sense that they
do not compute structure and motion directly from the images
but rather from a sparse set of precomputed feature points.
Despite a number of successes, they have several drawbacks:
• From the point of view of statistical inference, they are
suboptimal: In the selection of feature points much
potentially valuable information contained in the colors of
each image is discarded.
• They invariably lack robustness: Errors in the point
correspondence may have devastating effects on the
estimated camera motion. Since one often selects very
few point pairs only (8 points for the eight-point algorithm,
5 points for the five-point algorithm), any incorrect
correspondence will lead to an incorrect motion estimate.
• They do not address the highly coupled problems of
motion estimation and dense structure estimation. They
merely do so for a sparse set of points. As a consequence,
improvements in the estimated dense geometry will not be
used to improve the camera motion estimates.
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 5/33
Toward Direct Approaches to Multiview Reconstruction
In the last few years, researchers have been promoting direct
approaches to multi-view reconstruction. Rather than
extracting a sparse set of feature points to determine the
camera motion, direct methods aim at estimating camera
motion and dense or semi-dense scene geometry directly from
the input images. This has several advantages:
• Direct methods tend to be more robust to noise and other
nuisances because they exploit all available input
information.
• Direct methods provide a semi-dense geometric
reconstruction of the scene which goes well beyond the
sparse point cloud generated by the eight-point algorithm
or bundle adjustment. Depending on the application, a
separate dense reconstruction step may no longer be
necessary.
• Direct methods are typically faster because the
feature-point extraction and correspondence finding is
omitted: They can provide fairly accurate camera motion
and scene structure in real-time on a CPU.
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 6/33
Feature-Based versus Direct Methods
Engel, Sturm, Cremers, ICCV 2013
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 7/33
Direct Methods for Multi-view Reconstruction
In the following, we will briefly review several recent works on
direct methods for realtime multiple-view reconstruction:
• the method of Stühmer, Gumhold, Cremers, DAGM 2010
computes dense geometry from a handheld camera in
real-time.
• the methods of Steinbrücker, Sturm, Cremers, 2011 and
Kerl, Sturm, Cremers, 2013 directly compute the camera
motion of an RGB-D camera.
• the method of Newcombe, Lovegrove, Davison, ICCV
2011 directly determines dense geometry and camera
motion from the images.
• the method of Engel, Sturm, Cremers, ICCV 2013 and
Engel, Schöps, Cremers, ECCV 2014 directly computes
camera motion and semi-dense geometry for a handheld
(monocular) camera.
• the method of Engel, Koltun, Cremers, PAMI 2018 directly
estimates highly accurate camera motion and sparse
geometry.
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 8/33
Realtime Dense Geometry from a Handheld Camera
Let gi ∈ SE(3) be the rigid body motion from the first camera to
the i-th camera, and let Ii : Ω→ R be the i-th image. A dense
depth map h : Ω→ R can be computed by solving the
optimization problem:
min
h
n∑
i=2
∫
Ω
∣∣I1(x)− Ii(πgi (hx))∣∣ dx + λ ∫
Ω
|∇h|dx ,
where x is represented in homogeneous coordinates and hx is
the corresponding 3D point.
Like in optical flow estimation, the unknown depth map should
be such that for all pixels x ∈ Ω, the transformation into the
other images Ii should give rise to the same color as in the
reference image I1.
This cost function can be minimized at framerate by
coarse-to-fine linearization solved in parallel on a GPU.
Stuehmer, Gumhold, Cremers, DAGM 2010.
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 9/33
Realtime Dense Geometry from a Handheld Camera
Input image Reconstruction Textured geometry
Textured reconstructions Untextured
Stuehmer, Gumhold, Cremers, DAGM 2010.
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 10/33
Dense RGB-D Tracking
The approach of Stühmer et al. (2010) relies on a sparse
feature-point based camera tracker (PTAM) and computes
dense geometry directly on the images. Steinbrücker, Sturm,
Cremers (2011) propose a complementary approach to directly
compute the camera motion from RGB-D images. The idea is
to compute the rigid body motion gξ which optimally aligns two
subsequent color images I1 and I2:
min
ξ∈se(3)
∫
Ω
∣∣I1(x)− I2(πgξ(hx))∣∣2 dx
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 11/33
Dense RGB-D Tracking
The above non-convex problem can be approximated as a
convex problem by linearizing the residuum around an initial
guess ξ0:
E(ξ) ≈
∫
Ω
∣∣∣∣I1(x)− I2(πgξ0 (hx))−∇I>2
(
dπ
dgξ
)(
dgξ
dξ
)
ξ
∣∣∣∣2 dx
This is a convex quadratic cost function which gives rise to a
linear optimality condition:
dE(ξ)
dξ
= Aξ + b = 0
To account for larger motions of the camera, this problem is
solved in a coarse-to-fine manner. The linearization of the
residuum is identical with a Gauss-Newton approach. It
corresponds to an approximation of the Hessian by a positive
definite matrix.
Steinbrücker, Sturm, Cremers 2011
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 12/33
Dense RGB-D Tracking
Steinbrücker, Sturm, Cremers 2011
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 13/33
Dense RGB-D Tracking
In the small-baseline setting, this image aligning approach
provides more accurate camera motion than the commonly
used generalized Iterated Closest Points (GICP) approach.
frame difference, sequence 1 frame difference, sequence 2
Steinbrücker, Sturm, Cremers 2011
A related direct tracking approach was proposed for stereo
reconstruction in Comport, Malis, Rives, ICRA 2007. A
generalization which makes use of non-quadratic penalizers
was proposed in Kerl, Sturm, Cremers, ICRA 2013.
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 14/33
A Benchmark for RGB-D Tracking
Accurately tracking the camera is among the most central
challenges in computer vision. Quantitative performance of
algorithms can be validated on benchmarks.
Sturm, Engelhard, Endres, Burgard, Cremers, IROS 2012
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 15/33
A Benchmark for RGB-D Tracking
Sturm, Engelhard, Endres, Burgard, Cremers, IROS 2012
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 16/33
Combining Photometric and Geometric Consistency
Kerl, Sturm, Cremers, IROS 2013 propose an extension of the
RGB-D camera tracker which combines color consistency and
geometric consistency of subsequent RGB-D images.
Assuming that the vector ri = (rci , rzi ) ∈ R2 containing the color
and geometric discrepancy for pixel i follows a bivariate
t-distribution, the maximum likelihood pose estimate can be
computed as:
min
ξ∈R6
∑
i
wi r
>
i Σ
−1ri ,
with weights wi based on the student t-distribution:
wi =
ν + 1
ν + r>i Σ
−1ri
.
This nonlinear weighted least squares problem can be solved
in an iteratively reweighted least squares manner by alternating
a Gauss-Newton style optimization with a re-estimation of the
weights wi and the matrix Σ.
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 17/33
Loop Closure and Global Consistency
When tracking a camera over a longer period of time, errors
tend to accumulate. While a single room may still be mapped
more or less accurately, mapping a larger environment will lead
to increasing distortions: Corridors and walls will no longer be
straight but slightly curved.
A remedy is to introduce pose graph optimization and loop
closuring, a technique popularized in laser-based SLAM
systems. The key idea is to estimate the relative camera
motion ξ̂ij for any camera pair i and j in a certain neighborhood.
Subsequently, one can determine a globally consistent camera
trajectory ξ = {ξi}i=1..T by solving the nonlinear least squares
problem
min
ξ
∑
i∼j
(
ξ̂ij − ξi ◦ ξ−1j
)>
Σ−1ij
(
ξ̂ij − ξi ◦ ξ−1j
)
,
where Σ−1ij denotes the uncertainty of measurement ξ̂ij . This
problem can be solved using, for example, a
Levenberg-Marquardt algorithm.
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 18/33
Pose Graph Optimization and Loop Closure
Kerl, Sturm, Cremers, IROS 2013
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 19/33
Dense Tracking and Mapping
Newcombe, Lovegrove & Davison (ICCV 2011) propose an
algorithm which computes both the geometry of the scene and
the camera motion from a direct and dense algorithm.
They compute the inverse depth u = 1/h by minimizing a cost
function of the form
min
u
n∑
i=2
∫
Ω
∣∣∣I1(x)− Ii (πgi (xu
))∣∣∣ dx + λ ∫
Ω
ρ(x) |∇u|dx ,
for fixed camera motions gi . The function ρ introduces an
edge-dependent weighting assigning small weights in locations
where the input images exhibit strong gradients:
ρ(x) = exp (−|∇Iσ(x)|α) .
The camera tracking is then performed with respect to the
textured reconstruction in a manner similar to Steinbrücker et
al. (2011). The method is initialized using feature point based
stereo.
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 20/33
Dense Tracking and Mapping
Newcombe, Lovegrove & Davison (ICCV 2011)
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 21/33
Large-Scale Direct Monocular SLAM
A method for real-time direct monocular SLAM is proposed in
Engel, Sturm, Cremers, ICCV 2013 and Engel, Schöps,
Cremers, ECCV 2014. It combines several contributions which
make it well-suited for robust large-scale monocular SLAM:
• Rather than tracking and putting into correspondence a
sparse set of feature points, the method estimates a
semi-dense depth map which associates an inverse depth
with each pixel that exhibits sufficient gray value variation.
• To account for noise and uncertainty each inverse depth
value is associated with an uncertainty which is
propagated and updated over time like in a Kalman filter.
• Since monocular SLAM is invariably defined up to scale
only, we explicitly facilitate scaling of the reconstruction by
modeling the camera motion using the Lie group of 3D
similarity transformations Sim(3).
• Global consistency is assured by loop closuring on Sim(3).
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 22/33
Tracking by Direct sim(3) Image Alignment
Since reconstructions from a monocular camera are only
defined up to scale, Engel, Schöps, Cremers, ECCV 2014
account for rescaling of the environment by representing the
camera motion as an element in the Lie group of 3D similarity
transformations Sim(3) which is defined as:
Sim(3) =
{(
sR T
0 1
)
with R ∈ SO(3), T ∈ R3, s ∈ R+
}
.
One can minimize a nonlinear least squares problem
min
ξ∈sim(3)
∑
i
wi r
2
i (ξ),
where ri denotes the color residuum across different images
and wi a weighting as suggested in Kerl et al. IROS 2013.
The above cost function can then be optimized by a weighted
Gauss-Newton algorithm on the Lie group Sim(3):
ξ(t+1) = ∆ξ ◦ ξ(t), with ∆ξ =
(
J>WJ
)−1
J>W r , J =
∂r
∂ξ
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 23/33
Large-Scale Direct Monocular SLAM
Engel, Schöps, Cremers, ECCV 2014
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 24/33
Large-Scale Direct Monocular SLAM
Engel, Schöps, Cremers, ECCV 2014
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 25/33
Towards Direct Sparse Odometry
Despite its popularity, LSD SLAM has several shortcomings:
• While the pose graph optimization allows to impose global
consistency, it merely performs a joint optimization of the
extrinsic parameters associated with all keyframes. In
contrast to a full bundle adjustment, it does not optimize
the geometry. This is hard to do in realtime, in particular
for longer sequences.
• LSD SLAM actually optimizes two different cost functions
for estimating geometry and camera motion.
• LSD SLAM introduces spatial regularity by a spatial
filtering of the inverse depth values. This creates
correlations among the geometry parameters which in turn
makes Gauss-Newton optimization difficult.
• LSD SLAM is based on the assumption of brightness
constancy. In real-world videos, brightness is often not
preserved. Due to varying exposure time, vignette and
gamma correction, the brightness can vary substantially.
While feature descriptors are often invariant to these
changes, the local brightness itself is not.
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 26/33
From Brightness Constancy to Irradiance Constancy
Brightness variations due to vignette, gamma correction and
exposure time can be eliminated by a complete photometric
calibration:
I(x) = G
(
t V (x)B(x)
)
where the measured brightness I depends on the irradiance B,
the vignette V , the exposure time t and the camera response
function G (gamma function). G and V can be calibrated
beforehand, t can be read out from the camera.
Engel, Koltun, Cremers, PAMI 2018
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 27/33
Windowed Joint Optimization
A complete bundle adjustment over longer sequences is
difficult to carry out in realtime because the number of 3D point
coordinates may grow very fast over time. Furthermore new
observations are likely to predominantly affect parameters
associated with neighboring structures and cameras. For a
given data set, one can study the connectivity graph, i.e. a
graph where each node represents an image and two nodes
are connected if they look at the same 3D structure.
Direct Sparse Odometry therefore reverts to a windowed joint
optimization, the idea being that from all 3D coordinates and
camera frames only those in a recent time window are
included. The remaining ones are marginalized out.
If one avoids spatial filtering and selects only a sparser subset
of points, then the points can be assumed to be fairly
independent. As a result the Hessian matrix becomes sparser
and the Schur complement can be employed to make the
Gauss-Newton updates more efficient.
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 28/33
Effects of Spatial Correlation on the Hessian
Engel, Koltun, Cremers, PAMI 2018
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 29/33
Effect of Spatial Correlation on the Hessian Matrix
geometry not correlated geometry correlated
Engel, Koltun, Cremers, PAMI 2018
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 30/33
The Schur Complement Trick
Solving the Newton update step (called normal equation)
H x =
(
Hαα Hαβ
H>αβ Hββ
)(
xα
xβ
)
=
(
gα
gβ
)
,
for the unknowns xα and xβ is usually done by QR
decomposition for large problems. In this case, however, Hββ is
typically block diagonal (and thus easy to invert).
Left-multiplication with the matrix(
I −HαβH−1ββ
0 I
)
,
leads to:(
S 0
H>αβ Hββ
)(
xα
xβ
)
=
(
gα − HαβH−1ββ gβ
gβ
)
,
where S = Hαα −HαβH−1ββH
>
αβ is the Schur complement of Hββ
in H. It is symmetric, positive definite and block structured. The
equation Sxα = … is the reduced camera system.
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 31/33
Direct Sparse Odometry
Engel, Koltun, Cremers, PAMI 2018
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 32/33
Direct Sparse Odometry
Engel, Koltun, Cremers, PAMI 2018
Direct Approaches to
Visual SLAM
Prof. Daniel Cremers
Direct Methods
Realtime Dense
Geometry
Dense RGB-D
Tracking
Loop Closure and
Global Consistency
Dense Tracking and
Mapping
Large Scale Direct
Monocular SLAM
Direct Sparse
Odometry
updated April 12, 2021 33/33
Quantitative Evaluation
A quantitative comparison of Direct Sparse Odometry to the
state-of-the-art keypoint based technique ORB SLAM shows
substantial improvements in precision and robustness:
# of runs with a given error in translation, rotation and scale drift.
Engel, Koltun, Cremers, PAMI 2018
Direct Methods
Realtime Dense Geometry
Dense RGB-D Tracking
Loop Closure and Global Consistency
Dense Tracking and Mapping
Large Scale Direct Monocular SLAM
Direct Sparse Odometry