**Feature Non-Maximum Suppression**, or **FeatureNMS**, is a post-processing step for object detection models that removes duplicates where there are multiple detections outputted per object. FeatureNMS recognizes duplicates not only based on the intersection over union between the bounding boxes, but also based on the difference of feature vectors. These feature vectors can encode more information like visual appearance.

**Conditional Random Fields** or **CRFs** are a type of probabilistic graph model that take neighboring sample context into account for tasks like classification. Prediction is modeled as a graphical model, which implements dependencies between the predictions. Graph choice depends on the application, for example linear chain CRFs are popular in natural language processing, whereas in image-based tasks, the graph would connect to neighboring locations in an image to enforce that they have similar predictions.

Image Credit: [Charles Sutton and Andrew McCallum, An Introduction to Conditional Random Fields](https://homepages.inf.ed.ac.uk/csutton/publications/crftut-fnt.pdf)

FeatureNMS

FeatureNMS: Non-Maximum Suppression by Learning Feature Embeddings

**Inception-ResNet-v2** is a convolutional neural architecture that builds on the Inception family of architectures but incorporates [residual connections](https://paperswithcode.com/method/residual-connection) (replacing the filter concatenation stage of the Inception architecture).

Source	FeatureNMS: Non-Maximum Suppression by Learning Feature Embeddings
Year	2000
Data Source	CC BY-SA - https://paperswithcode.com