Title: Motion from image and inertial measurements additional slides
1Motion from image and inertial measurements
(additional slides)
- Dennis Strelow
- Carnegie Mellon University
2Outline
Robust image feature tracking (in detail)
Lucas-Kanade and real sequences The smalls
tracker Motion from omnidirectional images
3Robust image feature tracking Lucas-Kanade and
real sequences (1)
- Combining image and inertial measurements
improves our situation, but - we still need accurate feature tracking tracking
- some sequences do not come with inertial
measurements
4Robust image feature tracking Lucas-Kanade and
real sequences (2)
- better feature tracking for improved 6 DOF motion
estimation - remaining results will be image-only
5Robust image feature tracking Lucas-Kanade and
real sequences (3)
- Lucas-Kanade has been the go-to feature tracker
for shape-from-motion - minimizes a correlation-like matching error
- using general minimization
- evaluates the matching error at only a few
locations - subpixel resolution
6Robust image feature tracking Lucas-Kanade and
real sequences (4)
Additional heuristics used to apply Lucas-Kanade
to shape-from-motion
7Robust image feature tracking Lucas-Kanade and
real sequences (5)
But Lucas-Kanade performs poorly on many real
sequences
8Robust image feature tracking the smalls
tracker (1)
- smalls is a new feature tracker targeted at 6 DOF
motion estimation - exploits the rigid scene assumption
- eliminates the heuristics normally used with
Lucas-Kanade - SIFT is an enabling technology here
9Robust image feature tracking the smalls
tracker (2)
- First step epipolar geometry estimation
- use SIFT to establish matches between the two
images - get the 6 DOF camera motion between the two
images - get the epipolar geometry relating the two images
10Robust image feature tracking the smalls
tracker (3)
11Robust image feature tracking the smalls
tracker (4)
12Robust image feature tracking the smalls
tracker (5)
- Second step track along epipolar lines
- use nearby SIFT matches to get initial position
on epipolar line - exploits the rigid scene assumption
- eliminates heuristic pyramid
13Robust image feature tracking the smalls
tracker (6)
- Third step prune features
- geometrically inconsistent features are marked as
mistracked and removed - clumped features are pruned
- eliminates heuristic detecting mistracked
features based on convergence, error
14Robust image feature tracking the smalls
tracker (7)
- Fourth step extract new features
- spatial image coverage is the main criterion
- required texture is minimal when tracking is
restricted to the epipolar lines - eliminates heuristic extracting only textured
features -
15Robust image feature tracking the smalls
tracker (8)
16Robust image feature tracking the smalls
tracker (9)
left odometry only
right images only
- average error 1.74 m
- maximum error 5.14 m
17Robust image feature tracking the smalls
tracker (10)
- Recap
- exploits the rigid scene and eliminates
heuristics - allows hands-free tracking for real sequences
- can still be defeated by textureless areas or
repetitive texture
18Outline
Robust image feature tracking (in detail) Motion
from omnidirectional images
19Motion from omnidirectional images (1)
20Motion from omnidirectional images (2)
21Motion from omnidirectional images (3)
22Motion from omnidirectional images (4)
23Motion from omnidirectional images (5)
left non-rigid camera
right rigid camera
squares ground truth points solid
image-only estimates dash-dotted
image-and-inertial estimates
24Motion from omnidirectional images (6)
- In this experiment
- omni images
- conventional images inertial
- have roughly the same advantages
- But in general
- inertial has some advantages that omni images
alone cant produce - omni images can be harder to use