File size: 3,991 Bytes
c80917c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
"""
Preprocess a raw json dataset into features files for use in data_loader.py

Input: json file that has the form
[{ file_path: 'path/img.jpg', captions: ['a caption', ...] }, ...]
example element in this list would look like
{'captions': [u'A man with a red helmet on a small moped on a dirt road. ', u'Man riding a motor bike on a dirt road on the countryside.', u'A man riding on the back of a motorcycle.', u'A dirt path with a young person on a motor bike rests to the foreground of a verdant area with a bridge and a background of cloud-wreathed mountains. ', u'A man in a red shirt and a red hat is on a motorcycle on a hill side.'], 'file_path': u'val2014/COCO_val2014_000000391895.jpg', 'id': 391895}

This script reads this json, does some basic preprocessing on the captions
(e.g. lowercase, etc.), creates a special UNK token, and encodes everything to arrays

Output: two folders of features
"""

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

import os
import json
import argparse
from random import shuffle, seed
import string
# non-standard dependencies:
import h5py
from six.moves import cPickle
import numpy as np
import torch
import torchvision.models as models
import skimage.io

from torchvision import transforms as trn
preprocess = trn.Compose([
                #trn.ToTensor(),
                trn.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
])

from captioning.utils.resnet_utils import myResnet
import captioning.utils.resnet as resnet


def main(params):
    net = getattr(resnet, params['model'])()
    net.load_state_dict(torch.load(os.path.join(params['model_root'],params['model']+'.pth')))
    my_resnet = myResnet(net)
    my_resnet.cuda()
    my_resnet.eval()

    imgs = json.load(open(params['input_json'], 'r'))
    imgs = imgs['images']
    N = len(imgs)

    seed(123) # make reproducible

    dir_fc = params['output_dir']+'_fc'
    dir_att = params['output_dir']+'_att'
    if not os.path.isdir(dir_fc):
        os.mkdir(dir_fc)
    if not os.path.isdir(dir_att):
        os.mkdir(dir_att)

    for i,img in enumerate(imgs):
        # load the image
        I = skimage.io.imread(os.path.join(params['images_root'], img['filepath'], img['filename']))
        # handle grayscale input images
        if len(I.shape) == 2:
            I = I[:,:,np.newaxis]
            I = np.concatenate((I,I,I), axis=2)

        I = I.astype('float32')/255.0
        I = torch.from_numpy(I.transpose([2,0,1])).cuda()
        I = preprocess(I)
        with torch.no_grad():
            tmp_fc, tmp_att = my_resnet(I, params['att_size'])
        # write to pkl
        # print(dir_fc, str(img['cocoid']), tmp_fc.shape, tmp_att.shape, dir_att)
        # exit()
        np.save(os.path.join(dir_fc, str(img['cocoid'])), tmp_fc.data.cpu().float().numpy())
        np.savez_compressed(os.path.join(dir_att, str(img['cocoid'])), feat=tmp_att.data.cpu().float().numpy())

        if i % 1000 == 0:
            print('processing %d/%d (%.2f%% done)' % (i, N, i*100.0/N))
    print('wrote ', params['output_dir'])

if __name__ == "__main__":

    parser = argparse.ArgumentParser()

    # input json
    parser.add_argument('--input_json', required=True, help='input json file to process into hdf5')
    parser.add_argument('--output_dir', default='data', help='output h5 file')

    # options
    parser.add_argument('--images_root', default='', help='root location in which images are stored, to be prepended to file_path in input json')
    parser.add_argument('--att_size', default=14, type=int, help='14x14 or 7x7')
    parser.add_argument('--model', default='resnet101', type=str, help='resnet101, resnet152')
    parser.add_argument('--model_root', default='./data/imagenet_weights', type=str, help='model root')

    args = parser.parse_args()
    params = vars(args) # convert to ordinary dict
    print('parsed input parameters:')
    print(json.dumps(params, indent = 2))
    main(params)