search

zhaoweicai / mscnn

Caffe implementation of our multi-scale object detection framework
405 stars 211 forks source link

Question about the train/val/test sets #17

Closed MenglaiWang closed 8 years ago

MenglaiWang commented 8 years ago

Thanks for sharing your code! I have read some of source code. I am confused about your split of train/val/test sets. In the Caltech train_val.prototxt , training set is set00-05, the validation set is this image As shown in the figure, in the mscnn_window_file_caltech_test.txt , the validation images are from set06-10. So your testing set is the same as your validation set?? I wonder whether you have upload the wrong validation sets or I have a misunderstanding about Caltech pedestrian datasets. Hope for your reply.

zhaoweicai commented 8 years ago

Since the ground truth for Caltech testing set is available, we do not split the training set into train/val sets here. This follows other works on VOC2007 dataset. You can split the train/val sets by yourself.