Abstract:
We present an object recognition system that identifies objects by their silhouette from single views of a greyscale camera. The centroidal profile describing the object boundary is matched with boundaries from a model base using a dynamic programming technique. Objects are modelled by a multi-view representation which can either be learned from a set of images or generated from a geometric model. Experiments with non-rigid and polyhedral objects show the validity of this approach. 1. INTRODUCTION Vision-based object recognition is a field of research that has brought forth many systems that are able to recognise objects from single views of greyscale cameras, yet a fast recognition of a large number of free-form, perhaps even generic object classes in complex scenes with occlusion still seems to be a goal impossible to realise. The reason why we want to present another system that is clearly limited to a certain field of application is that we believe that the synthesis of various ...
Citations
|
1487
|
Dynamic programming
– Bellman
- 1957
|
|
744
|
Visual learning and recognition of 3-D objects from appearance
– Murase, Nayar
- 1995
|
|
128
|
On Image Analysis by the Methods of Moments
– The, Chin
- 1988
|
|
54
|
Application of affine-invariant fourier descriptors to recognition of 3-d objects
– Arbter, Snyder, et al.
- 1990
|
|
37
|
Recognition by functional parts
– Rivlin, Dickinson, et al.
- 1995
|
|
21
|
An Autoregressive Model Approach to Two-Dimensional Shape Classification
– Dubois, Glanz
- 1986
|
|
12
|
Learning to recognize objects in images: acquiring and using probabilistic models of appearance
– Pope
- 1995
|
|
11
|
A modification of deriche's approach to edge detection
– Lanser, Eckstein
- 1992
|
|
10
|
Architecture for computer vision application development within the horus system
– Eckstein, Steger
- 1997
|
|
10
|
A Hierarchical World Model with Sensor- and Task-Specific Features
– Hauck, Stoffler
- 1996
|
|
9
|
On the use of size functions for shape analysis
– Verri, Uras, et al.
- 1993
|
|
8
|
Aspect-Trees: Generation and Interpretation
– Munkelt
- 1995
|
|
4
|
Papadourakis Object recognition using invariant object boundary representations and neural network models
– Bebis, Papadourakis
- 1992
|
|
3
|
The challenge of generic object recognition
– Zerroug, Medioni
- 1994
|
|
2
|
Towards an Anthropomorphic Robotical Hand-Eye Coordination
– Fink, Hauck, et al.
- 1998
|
|
2
|
Hybrid Hand-Eye Coordination with a Single Stationary Camera
– Hauck, Farber
- 1996
|
|
2
|
Representations for recognizing complex curved 3d objects
– Kriegman, Ponce
- 1995
|
|
2
|
Paeseler A. `A Data-Driven Organisation of the Dynamic Programming Beam Search for Continuous Speech Recognition
– Ney, Mergel, et al.
- 1987
|
|
1
|
The Use of CharacteristicView Classes in the Recognition of Three-Dimensional Objects
– Chakravarty, Freeman
- 1980
|
|
1
|
Recognition and Localization of a 3D Polyhydral Object using a Neural Network
– Park, Cannon
- 1996
|
|
1
|
An algorithm for recognition and localisation of rotated and scaled objects
– Peli
- 1981
|
