aimagelab/show-control-and-tell

object_class_list.txt

Closed this issue · 2 comments

object_class_list.txt

excuse me,have you got the idea? thanks lot

Hi @miaoyi322 and @wangyuanshun,

For the moment, it is not possible to test our model on new images outside COCO and Flickr30k datasets.

Recall that, for each input image, our model takes as input a control signal in the form of a sequence or a set of image regions and predicts an output caption following the given control. In our experiments, we use the control signals coming from COCO and Flickr30k captions.