![]() ![]() Goodfellow, I.J., et al.: Generative adversarial networks. Geyer, C., Daniilidis, K.: A unifying theory for central panoramic systems and practical implications. Garcia-Garcia, A., Orts-Escolano, S., Oprea, S., Villena-Martinez, V., Rodríguez, J.G.: A review on deep learning techniques applied to semantic segmentation. ![]() In: Intelligent Vehicles Symposium (IV), pp. IEEE (2009)ĭeng, L., Yang, M., Qian, Y., Wang, C., Wang, B.: CNN based semantic segmentation for urban traffic scenes using fisheye camera. IEEE (2016)ĭeng, J., Dong, W., Socher, R., Li, L.J., Li, K., Fei-Fei, L.: ImageNet: a large-scale hierarchical image database. Ĭordts, M., et al.: The cityscapes dataset for semantic urban scene understanding. IEEE (2001)īrown, D.C.: Decentering distortion of lenses. Keywordsīarreto, J.P., Araujo, H.: Issues on the geometry of central catadioptric image formation. In this way we obtain a considerably higher semantic segmentation performance on the fisheye images: +18.3% intersection over union (IoU) for action-camera test images, +8.3% IoU for artificially generated fisheye data, and +18.0% IoU for challenging security scenes acquired in bird’s eye view. In this study, an alternative approach that modifies the augmentation stage of deep learning training to re-use rectilinear training data is presented. A potential solution to this problem is the recording and annotation of a new dataset, however this is expensive and tedious. Therefore, classical semantic segmentation approaches fall short in terms of performance compared to rectilinear data. ![]() The shape of objects is distorted depending on the distance between the principal point and the object position in the image. Semantic segmentation of fisheye images (e.g., from action-cameras or smartphones) requires different training approaches and data than those of rectilinear images obtained using central projection. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |