Counter-examples generation from a positive unlabeled image dataset

Chiaroni, Florent; Khodabandelou, Ghazaleh; Rahal, Mohamed-Cherif; Hueber, Nicolas; Dufaux, Frédéric

doi:10.1016/j.patcog.2020.107527

Cited by 14 publications

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 1 shows the results of our S2M sampling and the baselines for the non-overlapping classes. The performance of GenPU is reported only for MNIST-3/5 due to its mode collapse issue [39,40]. In the results, our S2M sampling adopted to unconditional GAN shows promising results in terms of both accuracy and FID † .…”

Section: Sampling For Classes With Inclusionmentioning

confidence: 99%

Mining Multi-Label Samples from Single Positive Labels

Cho¹,

Kim²,

Khan³

2022

Preprint

View full text Add to dashboard Cite

Conditional generative adversarial networks (cGANs) have shown superior results in class-conditional generation tasks. In order to simultaneously control multiple conditions, cGANs require multi-label training datasets, where multiple labels can be assigned to each data instance. Nevertheless, the tremendous annotation cost limits the accessibility of multi-label datasets in the real-world scenarios. Hence, we explore the practical setting called single positive setting, where each data instance is annotated by only one positive label with no explicit negative labels. To generate multi-label data in the single positive setting, we propose a novel sampling approach called single-to-multi-label (S2M) sampling, based on the Markov chain Monte Carlo method. As a widely applicable "add-on" method, our proposed S2M sampling enables existing unconditional and conditional GANs to draw high-quality multi-label data with a minimal annotation cost. Extensive experiments on real image datasets verify the effectiveness and correctness of our method, even when compared to a model trained with fully annotated datasets.* Equal contribution Preprint. Under review.

show abstract

Section: Sampling For Classes With Inclusionmentioning

confidence: 99%

Mining Multi-Label Samples from Single Positive Labels

Cho¹,

Kim²,

Khan³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The PU learning method with unique abilities has achieved great success in many fields 20 . For example, to overcome the issue of sensitivity to prior knowledge, Chiaroni et al 21 put forward a two‐stage GAN‐based PU learning method. Chen et al 22 proposed a cost‐effective multilabel active learning model which combined example‐label pair selection strategy and the most reliable positive sub‐example‐label pairs.…”

Section: Related Workmentioning

confidence: 99%

Positive unlabeled learning‐based anomaly detection in videos

Sun

Yuan

et al. 2021

Int J Intell Syst

View full text Add to dashboard Cite

Anomaly detection plays a critical role in intelligent video surveillance. However, real‐world video data obtained always contains large numbers of normal video data, along with large numbers of unlabeled data. A promising solution with one‐class classification and semi‐supervised learning may not be satisfactory as they fail to make good use of unlabeled data with only normal data available. In this paper, we introduced a new framework, called Positive Unlabeled learning‐based Anomaly event Detection (PU‐AD), to exploit the weakly‐supervised information. To the best of our knowledge, this is the first work that introduces the PU idea and achieves detecting abnormal events with a limited number of partially labeled data. Experiments on real‐world surveillance videos show that the proposed method outperforms the existing state‐of‐the‐art methods.

show abstract