SLIDE 1
An Object-based World Model and its Uses
Mac Mason Bhaskara Marthi
“I just want a robot that keeps track of my stuff”
- Aditi Nabar (spouse)
An Object-based World Model and its Uses Mac Mason Bhaskara Marthi - - PowerPoint PPT Presentation
An Object-based World Model and its Uses Mac Mason Bhaskara Marthi - - PowerPoint PPT Presentation
An Object-based World Model and its Uses Mac Mason Bhaskara Marthi I just want a robot that keeps track of my stuff - Aditi Nabar (spouse) Pipeline Point Planes in Plane Data Clouds frame Set of Kinect Plane Finder Association
SLIDE 2
SLIDE 3
SLIDE 4
Pipeline
Kinect Plane Finder Plane Data Association Clustering Object Data Association DB
Point Clouds Planes in frame Points above planes Object clusters Set of planes Set of objects + metadata
SLIDE 5
Implementation
- PR2 + Kinect
- ROS, PCL, Nav stack
- MongoDB for storing info
- Binary blobs + metadata (color, shape, ...)
- Partition by run
- Efficient queries over location, attributes
SLIDE 6
SLIDE 7
SLIDE 8
SLIDE 9
SLIDE 10
SLIDE 11
SLIDE 12
Data Association
- Match within runs based on spatial overlap
- Don’t commit to matches across runs
- Discover likely matches at query time
SLIDE 13
Attribute perception
- No object categories or instances
- Instead, extract descriptive features
Color: Red Shape: Cylinder Diameter: 0.1m
SLIDE 14
Sensors to symbols
- Sensor level
- Point clouds, images,
- Normals, SIFT features, ...
- Symbolic level: objects, categories, relations
- How to connect?
SLIDE 15
Bridging the gap
[Galindo et al, 2008]
SLIDE 16
Reify Geometry
- Surfaces, clusters, cluttered regions, ...
- Color, shape, topology, ...
- A lot can be done at this level!
SLIDE 17
Conclusions
- Object-based world model
- Allows
- Change detection
- Semantic querying
- Robust to noisy perception
SLIDE 18
Conclusions
- Object-based world model
- Allows
- Change detection
- Semantic querying
- Robust to noisy perception