[2002.04672] Object Detection as a Positive-Unlabeled Problemcontact arXivarXiv Twitter

As with other deep learning methods, label quality is important for learning modern convolutional object detectors. However, the potentially large number and wide diversity of object instances that can be found in complex image scenes makes constituting complete annotations a challenging task; objects missing annotations can be observed in a variety of popular object detection datasets. These missing annotations can be problematic, as the standard cross-entropy loss employed to train object detection models treats classification as a positive-negative (PN) problem: unlabeled regions are implicitly assumed to be background. As such, any object missing a bounding box results in a confusing learning signal, the effects of which we observe empirically. To remedy this, we propose treating object detection as a positive-unlabeled (PU) problem, which removes the assumption that unlabeled regions must be negative. We demonstrate that our proposed PU classification loss outperforms the standard

Date: 2020/02/14 20:21

Related Entries

Read more Google Cloud & NCAA® ML Competition 2020-NCAAM | Kaggle
0 users, 1 mentions 2020/02/18 23:21
Read more ALASKA2 Image Steganalysis | Kaggle
0 users, 1 mentions 2020/04/27 21:51
Read more [1607.07590] Simultaneous Estimation of Noise Variance and Number of Peaks in Bayesian Spectral Deco...
0 users, 1 mentions 2020/06/07 02:21
Read more SIIM-ISIC Melanoma Classification | Kaggle
0 users, 1 mentions 2020/08/09 02:21
Read more [2006.02595] Image Augmentations for GAN Trainingopen searchopen navigation menucontact arXivsubscri...
0 users, 1 mentions 2020/09/15 12:59