Distance Metric Between 3D Models and 2D Images for Recognition and Classification
IEEE Transactions on Pattern Analysis and Machine Intelligence
Extending case-based reasoning by discovering and using image features in IVF
SAC '00 Proceedings of the 2000 ACM symposium on Applied computing - Volume 1
| Hi-index | 0.01 |
Similarity between 3D objects and 2D images is typically measured using image metrics; namely, metrics that measure the difference in the image between the observed image and the nearest view of the object (e.g., the Euclidean distance between image points and their corresponding points in the nearest view). We introduce a different type of metrics: {\em transformation metrics}. These metrics penalize for the deformations applied to the object to produce the observed image. We present a metric that optimally penalizes for ``affine deformations'''' under weak- perspective. A closed-form solution is derived, and the metric is shown to bound the Euclidean image metric from both above and below