Domain Generalization With Domain Specific Aggregation Modules
The ability to build visual recognition algorithms able to work robustly in several visual domains without having access to any information about.
Domain generalization with domain specific aggregation modules. Domain generalization by solving jigsaw puzzles article carlucci2019domaingb title domain generalization by solving jigsaw puzzles author fabio maria carlucci and antonio d innocente and s. To reproduce results on pacs using alexnet on art painting split run. Submit remove a task domain generalization.
Tensorflow 1 15 for tensorboard logging reproducing results on pacs alexnet. The code was last tested with. We achieve this by introducing domain specific aggregation modules that through an aggregation layer strategy are able to merge generic and specific information in an effective manner.
D sam public repository for the paper domain generalization with domain specific aggregation modules. Experiments on two different benchmark databases show the power of our approach reaching the new state of the art in domain generalization.
The domain specific aggregation modules we propose can be seen as a way to augment the generalization abilities of given cnn architectures by maintaining a generic core while at the same time explicitly modeling the single domain specific features separately in a whole coherent structure. Request pdf domain generalization with domain specific aggregation modules visual recognition systems are meant to work in the real world. Bucci and barbara caputo and t.
40th german conference gcpr 2018 stuttgart germany october 9 12 2018 proceedings visual recognition systems. Parent task if any. Within this context a very realistic scenario deals with domain generalization i e.
Request pdf domain generalization with domain specific aggregation modules. We achieve this by introducing domain specific aggregation modules that through an aggregation layer strategy are able to merge generic and specific information in an effective manner. Create a new task.