A Computational Approach to Edge Detection
IEEE Transactions on Pattern Analysis and Machine Intelligence
IEEE Transactions on Pattern Analysis and Machine Intelligence
A Performance Evaluation of Local Descriptors
IEEE Transactions on Pattern Analysis and Machine Intelligence
Speeded-Up Robust Features (SURF)
Computer Vision and Image Understanding
Performance evaluation of local colour invariants
Computer Vision and Image Understanding
PCA-SIFT: a more distinctive representation for local image descriptors
CVPR'04 Proceedings of the 2004 IEEE computer society conference on Computer vision and pattern recognition
| Hi-index | 0.00 |
Statistical patch-based observation (SPBO) is built specifically for obtaining good tracking observation in robust environment. In video analytics applications, the problems of blurring, moderate deformation, low ambient illumination, homogenous texture and illumination change are normally encountered as the foreground objects move. We approach the problems by fusing both feature and template based methods. While we believe that feature based matchings are more distinctive, we consider that object matching is best achieved by means of a collection of points as in template based detectors. Our algorithm starts by building comparison vectors at each detected point of interest between consecutive frames. The vectors are matched to build possible patches based on their respective coordination. Patch matching is done statistically by modelling the histograms of patches as Poisson distributions for both RGB and HSV colour models. Then, maximum likelihood is applied for position smoothing while a Bayesian approach is applied for size smoothing. Our algorithm performs better than SIFT and SURF detectors in a majority of the cases especially in complex video scenes.