Hello All,
I am training my system against Partially Annotated Databases The Caltech Database given here http://pascallin.ecs.soton.ac.uk/challenges/VOC/databases.html.
I am using "bag of words" algorithm for vocabulary building and detection. My YML file is huge. So everytime I pass an image for detection the ~65MB yml file is loaded and then the object is detected. How can I do it more efficiently ?? Or is there some other way to store the vocabulary ??
I am badly stuck at it please help!!
Thanks in advance. Aditi K