Loading...
Please wait, while we are loading the content...
Similar Documents
Object detection with large intra-class variation
| Content Provider | Semantic Scholar |
|---|---|
| Author | Chen, Guang |
| Copyright Year | 2011 |
| Abstract | For object detection, the state-of-the-art performance is achieved through supervised learning. The performances of object detectors of this kind are mainly determined by two factors: features and underlying classif cation algorithms. In this work, we aim at improving the performance of object detectors from the aspect of classif cation algorithm. Observing the fact that classif ers used for object detection are task dependent and data driven, we developed a hybrid learning algorithm combining global classif cation and local adaptations, which automatically adjusts model complexity according to data distribution. We divide data samples into two groups, easy samples and ambiguous samples, using a learned global classif er. A local adaptation approach based on spectral clustering and propsed Min-Max model adaptation is then applied to further process the ambiguous samples. The proposed algorithm automatically determines model complexity of the local learning algorithm according to the distribution of ambiguous samples. By autonomously striking a balance between model complexity and learning capacity, the proposed hybrid learning algorithm incarnates a human detector outperforming the state-of-the-art algorithms on a couple of benchmark datasets [2, 1] and a self-collected pedestrian dataset. Besides , the proposed Min-Max model adaptation algorithm also successfully improve the performance of an off ine-trained classif er on-site by adapting the classif er towards newly acquired data, without worries about the tuning the adaptation rate parameter, which affects the performance gain substantially. Taking the object detection as a testbed, we implement an adapted object detector based on binary classif cation. Under different adaptation scenarios and different datasets including PASCAL, ImageNet, INRIA, and TUD-Pedestrian, the proposed adaption method achieves signif cant performance gain and is compared favorably with the state-of-the-art adaptation method with the f ne tuned adaptation rate. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | https://mospace.umsystem.edu/xmlui/bitstream/handle/10355/14534/research.pdf?isAllowed=y&sequence=2 |
| Alternate Webpage(s) | https://mospace.umsystem.edu/xmlui/bitstream/handle/10355/14534/short.pdf?isAllowed=y&sequence=3 |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
