Two Video Data Sets for Tracking and Retrieval of Out of Distribution Objects

Maag, Kira; Chan, Robin; Svenja, Uhlemeyer,; Kowol, Kamil; Gottschalk, Hanno

doi:10.1007/978-3-031-26348-4_28

Cited by 3 publications

(1 citation statement)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figure 5 ). In parallel, significant effort has been invested into artificial datasets by leveraging simulated environments [ 59 , 78 ]. Similarly, remote-sensing segmentation datasets have grown both in size in complexity [ 16 ].…”

Section: Experimental Setupmentioning

confidence: 99%

Dense Out-of-Distribution Detection by Robust Learning on Synthetic Negative Data

Grcić,

Bevandić,

Kalafatić

et al. 2024

Sensors

View full text Add to dashboard Cite

Standard machine learning is unable to accommodate inputs which do not belong to the training distribution. The resulting models often give rise to confident incorrect predictions which may lead to devastating consequences. This problem is especially demanding in the context of dense prediction since input images may be only partially anomalous. Previous work has addressed dense out-of-distribution detection by discriminative training with respect to off-the-shelf negative datasets. However, real negative data may lead to over-optimistic evaluation due to possible overlap with test anomalies. To this end, we extend this approach by generating synthetic negative patches along the border of the inlier manifold. We leverage a jointly trained normalizing flow due to a coverage-oriented learning objective and the capability to generate samples at different resolutions. We detect anomalies according to a principled information-theoretic criterion which can be consistently applied through training and inference. The resulting models set the new state of the art on benchmarks for out-of-distribution detection in road-driving scenes and remote sensing imagery despite minimal computational overhead.

show abstract

Section: Experimental Setupmentioning

confidence: 99%