This repository contains the code and the data for the paper "Text Attribute Aggregation and Visual Feature Decomposition for Person Search"
Iodice, Sara and Mikolajczyk, Krystian
Imperial College London, London
Download attributes/annotation, and pretrained model from here and unzip its content into "data" folder.
Download Market-1501 dataset from here and unzip it in the "data" folder.
To train the model run the script training_script.sh
To test the model run the script testing_script.sh
If you have any questions about this code, contact us.