DenseHybrid: Hybrid Anomaly Detection for Dense Open-set Recognition

by   Matej Grcić, et al.

Anomaly detection can be conceived either through generative modelling of regular training data or by discriminating with respect to negative training data. These two approaches exhibit different failure modes. Consequently, hybrid algorithms present an attractive research goal. Unfortunately, dense anomaly detection requires translational equivariance and very large input resolutions. These requirements disqualify all previous hybrid approaches to the best of our knowledge. We therefore design a novel hybrid algorithm based on reinterpreting discriminative logits as a logarithm of the unnormalized joint distribution p̂(𝐱, 𝐲). Our model builds on a shared convolutional representation from which we recover three dense predictions: i) the closed-set class posterior P(𝐲|𝐱), ii) the dataset posterior P(d_in|𝐱), iii) unnormalized data likelihood p̂(𝐱). The latter two predictions are trained both on the standard training data and on a generic negative dataset. We blend these two predictions into a hybrid anomaly score which allows dense open-set recognition on large natural images. We carefully design a custom loss for the data likelihood in order to avoid backpropagation through the untractable normalizing constant Z(θ). Experiments evaluate our contributions on standard dense anomaly detection benchmarks as well as in terms of open-mIoU - a novel metric for dense open-set performance. Our submissions achieve state-of-the-art performance despite neglectable computational overhead over the standard semantic segmentation baseline.


Hybrid Open-set Segmentation with Synthetic Negative Data

Open-set segmentation is often conceived by complementing closed-set cla...

Real time dense anomaly detection by learning on synthetic negative data

Most approaches to dense anomaly detection rely on generative modeling o...

Dense anomaly detection by robust learning on synthetic negative data

Standard machine learning is unable to accommodate inputs which do not b...

On advantages of Mask-level Recognition for Open-set Segmentation in the Wild

Most dense recognition methods bring a separate decision in each particu...

Dense outlier detection and open-set recognition based on training with noisy negative images

Deep convolutional models often produce inadequate predictions for input...

Env-Aware Anomaly Detection: Ignore Style Changes, Stay True to Content!

We introduce a formalization and benchmark for the unsupervised anomaly ...

Dense open-set recognition with synthetic outliers generated by Real NVP

Today's deep models are often unable to detect inputs which do not belon...

Please sign up or login with your details

Forgot password? Click here to reset