Source-Free-Object-Detection-by-Learning-to-Overlook-Domain-Style
Source-Free-Object-Detection-by-Learning-to-Overlook-Domain-Style copied to clipboard
Published
20 hours ago •
Flashkong
Flashkong
About structure of data sets folders
How do you organize the structure of data sets, and how do you organize the naming of training and testing data set folders, such as in "Prepare Data" ,what is the "/home/lishuaifeng/data/clipart/VOC2007/ImageSets/Main/traintest1k.txt" ? It is not in the PASCAL VOC dataset.
We want more details about the the structure of data sets folders, so as to train the model
I am very sorry for the long delay in replying, and we really appreciate your interest in our papers!
/home/lishuaifeng/data/clipart/VOC2007/ImageSets/Main/traintest1k.txt is a collection of training and test data of the clipart dataset. It has 1000 images.
The structure of the dataset is as follows：
Cityscape
└── VOC2007
├── Annotations
├── JPEGImages
├── ImageSets
└── Main
├── test_all500.txt # no foggy 500 test images and foggy 500 test images
├── test_all.txt # no foggy 347 test images and foggy 500 test images
├── test_s500.txt # no foggy test, 500 images (KITTI->city, we use this to test the LODS model)
├── test_s.txt # no foggy test, 347 images
├── test_t.txt # foggy test 500 images (city->foggy, we use this to test the LODS model)
├── train_all.txt # no foggy and foggy train images
├── train_s.txt # no foggy train images (city->foggy, we use this to train the source only model. KITTI->city, we use this to train the LODS model)
└── train_t.txt # foggy train images (city->foggy, we use this to train the LODS model)
Clipart
└── VOC2007
├── Annotations
├── JPEGImages
└── ImageSets
└── Main
├── test.txt # 500 test images
├── traintest1k.txt # 1000 images used when train and test (we use this to train and test LODS model)
└── train.txt # 500 train images
Watercolor
└── VOC2007
├── Annotations
├── JPEGImages
└── ImageSets
└── Main
├── all.txt # all images in the dataset (containing unlabeled)
├── extra.txt # images which is unlabeled
├── instance_level_annotated.txt # 2000 labeled images
├── test_wc.txt # 1000 test images (we use this to test LODS model)
└── train_wc.txt # 1000 train images (we use this to train LODS model)
KITTI
└── VOC2007
├── Annotations
├── JPEGImages
└── ImageSets
├── Layout
├── Main
│ ├── car_test.txt
│ ├── car_train500.txt
│ ├── car_trainall.txt
│ ├── car_val.txt
│ ├── pedestrian_test.txt
│ ├── pedestrian_train500.txt
│ ├── pedestrian_trainall.txt
│ ├── pedestrian_val.txt
│ ├── test.txt
│ ├── train500.txt
│ ├── trainall.txt # 7481 images (for the KITTI -> cityscape, we use this to train the source only model)
│ ├── tram_test.txt
│ ├── tram_train500.txt
│ ├── tram_trainall.txt
│ ├── tram_val.txt
│ ├── truck_test.txt
│ ├── truck_train500.txt
│ ├── truck_trainall.txt
│ ├── truck_val.txt
│ ├── val.txt
│ ├── van_test.txt
│ ├── van_train500.txt
│ ├── van_trainall.txt
│ └── van_val.txt
└── Segmentation
VOCdevkit
├── VOC2007
│ ├── Annotations
│ ├── JPEGImages
│ ├── SegmentationClass
│ ├── SegmentationObject
│ └── ImageSets
│ ├── Layout
│ │ ├── test.txt
│ │ ├── train.txt
│ │ ├── trainval.txt
│ │ └── val.txt
│ ├── Main
│ │ ├── aeroplane_test.txt
│ │ ├── aeroplane_train.txt
│ │ ├── aeroplane_trainval.txt
│ │ ├── aeroplane_val.txt
│ │ ├── bicycle_test.txt
│ │ ├── bicycle_train.txt
│ │ ├── bicycle_trainval.txt
│ │ ├── bicycle_val.txt
│ │ ├── bird_test.txt
│ │ ├── bird_train.txt
│ │ ├── bird_trainval.txt
│ │ ├── bird_val.txt
│ │ ├── boat_test.txt
│ │ ├── boat_train.txt
│ │ ├── boat_trainval.txt
│ │ ├── boat_val.txt
│ │ ├── bottle_test.txt
│ │ ├── bottle_train.txt
│ │ ├── bottle_trainval.txt
│ │ ├── bottle_val.txt
│ │ ├── bus_test.txt
│ │ ├── bus_train.txt
│ │ ├── bus_trainval.txt
│ │ ├── bus_val.txt
│ │ ├── car_test.txt
│ │ ├── car_train.txt
│ │ ├── car_trainval.txt
│ │ ├── car_val.txt
│ │ ├── cat_test.txt
│ │ ├── cat_train.txt
│ │ ├── cat_trainval.txt
│ │ ├── cat_val.txt
│ │ ├── chair_test.txt
│ │ ├── chair_train.txt
│ │ ├── chair_trainval.txt
│ │ ├── chair_val.txt
│ │ ├── cow_test.txt
│ │ ├── cow_train.txt
│ │ ├── cow_trainval.txt
│ │ ├── cow_val.txt
│ │ ├── diningtable_test.txt
│ │ ├── diningtable_train.txt
│ │ ├── diningtable_trainval.txt
│ │ ├── diningtable_val.txt
│ │ ├── dog_test.txt
│ │ ├── dog_train.txt
│ │ ├── dog_trainval.txt
│ │ ├── dog_val.txt
│ │ ├── horse_test.txt
│ │ ├── horse_train.txt
│ │ ├── horse_trainval.txt
│ │ ├── horse_val.txt
│ │ ├── motorbike_test.txt
│ │ ├── motorbike_train.txt
│ │ ├── motorbike_trainval.txt
│ │ ├── motorbike_val.txt
│ │ ├── person_test.txt
│ │ ├── person_train.txt
│ │ ├── person_trainval.txt
│ │ ├── person_val.txt
│ │ ├── pottedplant_test.txt
│ │ ├── pottedplant_train.txt
│ │ ├── pottedplant_trainval.txt
│ │ ├── pottedplant_val.txt
│ │ ├── sheep_test.txt
│ │ ├── sheep_train.txt
│ │ ├── sheep_trainval.txt
│ │ ├── sheep_val.txt
│ │ ├── sofa_test.txt
│ │ ├── sofa_train.txt
│ │ ├── sofa_trainval.txt
│ │ ├── sofa_val.txt
│ │ ├── test.txt
│ │ ├── train_test.txt
│ │ ├── train_train.txt
│ │ ├── train_trainval.txt
│ │ ├── train.txt
│ │ ├── train_val.txt
│ │ ├── trainval.txt # 5011 images (pascal voc -> clipart and pascal voc -> watercolor, we use this train the source only model)
│ │ ├── tvmonitor_test.txt
│ │ ├── tvmonitor_train.txt
│ │ ├── tvmonitor_trainval.txt
│ │ ├── tvmonitor_val.txt
│ │ └── val.txt
│ └── Segmentation
│ ├── test.txt
│ ├── train.txt
│ ├── trainval.txt
│ └── val.txt
└── VOC2012
├── Annotations
├── JPEGImages
├── SegmentationClass
├── SegmentationObject
└── ImageSets
├── Action
│ ├── jumping_test.txt
│ ├── jumping_train.txt
│ ├── jumping_trainval.txt
│ ├── jumping_val.txt
│ ├── phoning_test.txt
│ ├── phoning_train.txt
│ ├── phoning_trainval.txt
│ ├── phoning_val.txt
│ ├── playinginstrument_test.txt
│ ├── playinginstrument_train.txt
│ ├── playinginstrument_trainval.txt
│ ├── playinginstrument_val.txt
│ ├── reading_test.txt
│ ├── reading_train.txt
│ ├── reading_trainval.txt
│ ├── reading_val.txt
│ ├── ridingbike_test.txt
│ ├── ridingbike_train.txt
│ ├── ridingbike_trainval.txt
│ ├── ridingbike_val.txt
│ ├── ridinghorse_test.txt
│ ├── ridinghorse_train.txt
│ ├── ridinghorse_trainval.txt
│ ├── ridinghorse_val.txt
│ ├── running_test.txt
│ ├── running_train.txt
│ ├── running_trainval.txt
│ ├── running_val.txt
│ ├── takingphoto_test.txt
│ ├── takingphoto_train.txt
│ ├── takingphoto_trainval.txt
│ ├── takingphoto_val.txt
│ ├── test.txt
│ ├── train.txt
│ ├── trainval.txt
│ ├── usingcomputer_test.txt
│ ├── usingcomputer_train.txt
│ ├── usingcomputer_trainval.txt
│ ├── usingcomputer_val.txt
│ ├── val.txt
│ ├── walking_test.txt
│ ├── walking_train.txt
│ ├── walking_trainval.txt
│ └── walking_val.txt
├── Layout
│ ├── test.txt
│ ├── train.txt
│ ├── trainval.txt
│ └── val.txt
├── Main
│ ├── aeroplane_test.txt
│ ├── aeroplane_train.txt
│ ├── aeroplane_trainval.txt
│ ├── aeroplane_val.txt
│ ├── bicycle_test.txt
│ ├── bicycle_train.txt
│ ├── bicycle_trainval.txt
│ ├── bicycle_val.txt
│ ├── bird_test.txt
│ ├── bird_train.txt
│ ├── bird_trainval.txt
│ ├── bird_val.txt
│ ├── boat_test.txt
│ ├── boat_train.txt
│ ├── boat_trainval.txt
│ ├── boat_val.txt
│ ├── bottle_test.txt
│ ├── bottle_train.txt
│ ├── bottle_trainval.txt
│ ├── bottle_val.txt
│ ├── bus_test.txt
│ ├── bus_train.txt
│ ├── bus_trainval.txt
│ ├── bus_val.txt
│ ├── car_test.txt
│ ├── car_train.txt
│ ├── car_trainval.txt
│ ├── car_val.txt
│ ├── cat_test.txt
│ ├── cat_train.txt
│ ├── cat_trainval.txt
│ ├── cat_val.txt
│ ├── chair_test.txt
│ ├── chair_train.txt
│ ├── chair_trainval.txt
│ ├── chair_val.txt
│ ├── cow_test.txt
│ ├── cow_train.txt
│ ├── cow_trainval.txt
│ ├── cow_val.txt
│ ├── diningtable_test.txt
│ ├── diningtable_train.txt
│ ├── diningtable_trainval.txt
│ ├── diningtable_val.txt
│ ├── dog_test.txt
│ ├── dog_train.txt
│ ├── dog_trainval.txt
│ ├── dog_val.txt
│ ├── horse_test.txt
│ ├── horse_train.txt
│ ├── horse_trainval.txt
│ ├── horse_val.txt
│ ├── motorbike_test.txt
│ ├── motorbike_train.txt
│ ├── motorbike_trainval.txt
│ ├── motorbike_val.txt
│ ├── person_test.txt
│ ├── person_train.txt
│ ├── person_trainval.txt
│ ├── person_val.txt
│ ├── pottedplant_test.txt
│ ├── pottedplant_train.txt
│ ├── pottedplant_trainval.txt
│ ├── pottedplant_val.txt
│ ├── sheep_test.txt
│ ├── sheep_train.txt
│ ├── sheep_trainval.txt
│ ├── sheep_val.txt
│ ├── sofa_test.txt
│ ├── sofa_train.txt
│ ├── sofa_trainval.txt
│ ├── sofa_val.txt
│ ├── test.txt
│ ├── train_test.txt
│ ├── train_train.txt
│ ├── train_trainval.txt
│ ├── train.txt
│ ├── train_val.txt
│ ├── trainval.txt # 11540 images (pascal voc -> clipart and pascal voc -> watercolor, we use this train the source only model)
│ ├── tvmonitor_test.txt
│ ├── tvmonitor_train.txt
│ ├── tvmonitor_trainval.txt
│ ├── tvmonitor_val.txt
│ └── val.txt
└── Segmentation
├── test.txt
├── train.txt
├── trainval.txt
└── val.txt
