Spatial Location Constraint Prototype Loss for Open Set Recognition
This work addresses the challenge of identifying unknown classes in pattern recognition, which is crucial for real-world applications like security and autonomous systems, though it appears incremental as it builds on existing methods.
The paper tackles open set recognition by proposing a spatial location constraint prototype loss to simultaneously reduce empirical and open space risks, achieving superior performance on multiple benchmark datasets compared to existing approaches.
One of the challenges in pattern recognition is open set recognition. Compared with closed set recognition, open set recognition needs to reduce not only the empirical risk, but also the open space risk, and the reduction of these two risks corresponds to classifying the known classes and identifying the unknown classes respectively. How to reduce the open space risk is the key of open set recognition. This paper explores the origin of the open space risk by analyzing the distribution of known and unknown classes features. On this basis, the spatial location constraint prototype loss function is proposed to reduce the two risks simultaneously. Extensive experiments on multiple benchmark datasets and many visualization results indicate that our methods is superior to most existing approaches.