pith. sign in

arxiv: 1901.08394 · v1 · pith:VN5ELP5Nnew · submitted 2019-01-24 · 💻 cs.CV · cs.LG· stat.ML

Application of Decision Rules for Handling Class Imbalance in Semantic Segmentation

classification 💻 cs.CV cs.LGstat.ML
keywords classclassesdatasettrainingdataimbalanceinstancesnetwork
0
0 comments X
read the original abstract

As part of autonomous car driving systems, semantic segmentation is an essential component to obtain a full understanding of the car's environment. One difficulty, that occurs while training neural networks for this purpose, is class imbalance of training data. Consequently, a neural network trained on unbalanced data in combination with maximum a-posteriori classification may easily ignore classes that are rare in terms of their frequency in the dataset. However, these classes are often of highest interest. We approach such potential misclassifications by weighting the posterior class probabilities with the prior class probabilities which in our case are the inverse frequencies of the corresponding classes in the training dataset. More precisely, we adopt a localized method by computing the priors pixel-wise such that the impact can be analyzed at pixel level as well. In our experiments, we train one network from scratch using a proprietary dataset containing 20,000 annotated frames of video sequences recorded from street scenes. The evaluation on our test set shows an increase of average recall with regard to instances of pedestrians and info signs by $25\%$ and $23.4\%$, respectively. In addition, we significantly reduce the non-detection rate for instances of the same classes by $61\%$ and $38\%$.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. The Ethical Dilemma when (not) Setting up Cost-based Decision Rules in Semantic Segmentation

    cs.CV 2019-07 unverdicted novelty 5.0

    Defining egoistic and altruistic cost functions for class confusions in semantic segmentation changes precision, recall, and segment-wise error rates relative to standard MAP decisions.

  2. A Comparative Study of High-Recall Real-Time Semantic Segmentation Based on Swift Factorized Network

    cs.CV 2019-07 unverdicted novelty 4.0

    Adapts SwiftNet into SFN with ERFNet/GCNet-inspired blocks and compares loss functions, classifiers, and decision rules to increase recall in real-time semantic segmentation on CamVid and Cityscapes.

  3. Importance-Aware Semantic Segmentation with Efficient Pyramidal Context Network for Navigational Assistant Systems

    cs.CV 2019-07 unverdicted novelty 3.0

    Introduces importance-aware loss and BiERF-PSPNet extension for semantic segmentation tailored to navigational assistant systems, evaluated on CamVid and Cityscapes.