1. Place all the data in the following structure, and there are 130 scenes in total.

|-- google1000
    |-- scenes/
    |   |--- scene_0000/
    |   |--- scene_0001/
    |   |--- ... ...
    |   |--- scene_1499/
    |
    |-- models
    |   |-- 000/                        # Details of model of object 0
    |   |-- ... ...
    |   `-- 1029/                        
    |-- models_down
    |   |-- 000.ply                     # Downsampled model point cloudof object 0
    |   |-- ... ...
	|   `-- 1029.ply   
    |
    |-- camera.json                     # Camera intrinsics
    |
    `-- graspnet_labels_v3              # correspondence lables between scenes and object models in ./models_down/; initially empty, need further labelling


2. Detail structure of each scene (take scene_0000 as an example)
|-- scene_0000
    |-- blender_proc                    
    |   |   |-- rgb                         
    |   |   |   |-- 0000.jpg to 049.jpg    # 50 rgb images
    |   |   `-- depth
    |   |   |   |-- 0000.png to 049.png    # 50 depth images
    |   |   `-- label
    |   |   |   |-- 0000.png to 049.png    # 50 object mask images, 0 is background, 1-88 denotes each object (1-indexed), same format as YCB-Video dataset
    |   |   `-- annotations
    |   |   |   |-- 0000.xml to 049.xml    # 50 object 6d pose annotation. 'pos_in_world' and'ori_in_world' denotes position and orientation w.r.t the camera frame. 
    |   |   `-- meta
    |   |   |   |-- 0000.mat to 049.mat    # 50 object 6d pose annotation, same format as YCB-Video dataset for easy usage
    |   |   `-- camK.npy                   # camera intrinsic, shape: 3x3, [[f_x,0,c_x], [0,f_y,c_y], [0,0,1]]
    |   |   `-- camera_poses.npy           # 50 camera poses with respect to the first frame, shape: 256x(4x4)
    |   |   `-- cam0_wrt_table.npy         # first frame's camera pose with respect to the table, shape: 4x4
	|
    `-- object_id_list.txt              # ids of objects appeared in this scene