Scene Novelty Prediction from Unsupervised Discriminative Feature Learning
Paper in proceeding, 2020
Our approach to gain this additional information is to estimate how similar test data is to the training data that the model was trained on. We map training instances onto a feature space that is the most discriminative among them. We then model the entire training set as a Gaussian distribution in that feature space. The novelty of the test data is characterized by its low probability of being in that distribution, or equivalently a large Mahalanobis distance in the feature space.
Our distance metric in the discriminative feature space achieves a better novelty prediction performance than the state-of-the-art methods on most classes in CIFAR-10 and ImageNet. Using semantic segmentation as a proxy task often needed for autonomous driving, we show that our unsupervised novelty prediction correlates with the performance of a segmentation network trained on full pixel-wise annotations. These experimental results demonstrate potential applications of our method upon identifying scene familiarity and quantifying the confidence in autonomous driving actions.
Author
Arian Ranjbar
Chalmers, Electrical Engineering, Systems and control
Chun-Hsiao Yeh
University of California at Berkeley
Sascha Hornauer
University of California at Berkeley
Stella Yu
University of California at Berkeley
Ching-Yao Chan
University of California at Berkeley
2020 IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC)
, ,
Subject Categories
Other Computer and Information Science
Bioinformatics (Computational Biology)
Computer Science