BSN-boundary-sensitive-network/TEM_load_data.py at master · wzmsltw/BSN-boundary-sensitive-network · GitHub

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
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
import random
import numpy as np
import pandas as pd
import json

tscale = 100
tgap = 1. / tscale

def load_json(file):
    with open(file) as json_file:
        data = json.load(json_file)
        return data

def getDatasetDict():
    """Load dataset file
    """
    df=pd.read_csv("./data/activitynet_annotations/video_info_new.csv")
    json_data= load_json("./data/activitynet_annotations/anet_anno_action.json")
    database=json_data
    train_dict={}
    val_dict={}
    test_dict={}
    for i in range(len(df)):
        video_name=df.video.values[i]
        video_info=database[video_name]
        video_new_info={}
        video_new_info['duration_frame']=video_info['duration_frame']
        video_new_info['duration_second']=video_info['duration_second']
        video_new_info["feature_frame"]=video_info['feature_frame']
        video_subset=df.subset.values[i]
        video_new_info['annotations']=video_info['annotations']
        if video_subset=="training":
            train_dict[video_name]=video_new_info
        elif video_subset=="validation":
            val_dict[video_name]=video_new_info
        elif video_subset=="testing":
            test_dict[video_name]=video_new_info
    return train_dict,val_dict,test_dict

def ioa_with_anchors(anchors_min,anchors_max,box_min,box_max):
    """Compute intersection between score a box and the anchors.
    """
    len_anchors=anchors_max-anchors_min
    int_xmin = np.maximum(anchors_min, box_min)
    int_xmax = np.minimum(anchors_max, box_max)
    inter_len = np.maximum(int_xmax - int_xmin, 0.)
    scores = np.divide(inter_len, len_anchors)
    return scores

def getBatchList(numVideo,batch_size,shuffle=True):
    """Generate batch list for each epoch randomly
    """
    video_list=range(numVideo)
    batch_start_list=[i*batch_size for i in range(len(video_list)/batch_size)]
    batch_start_list.append(len(video_list)-batch_size)
    if shuffle==True:
        random.shuffle(video_list)
    batch_video_list=[]
    for bstart in batch_start_list:
        batch_video_list.append(video_list[bstart:(bstart+batch_size)])
    return batch_video_list

def getBatchListTest(video_dict,batch_size,shuffle=True):
    """Generate batch list during testing
    """
    video_list=video_dict.keys()
    batch_start_list=[i*batch_size for i in range(len(video_list)/batch_size)]
    batch_start_list.append(len(video_list)-batch_size)
    if shuffle==True:
        random.shuffle(video_list)
    batch_video_list=[]
    for bstart in batch_start_list:
        batch_video_list.append(video_list[bstart:(bstart+batch_size)])
    return batch_video_list

def getBatchData(video_list,data_dict):
    """Given a video list (batch), get corresponding data
    """
    batch_label_action=[]
    batch_label_start=[]
    batch_label_end=[]
    batch_anchor_feature=[]

    for idx in video_list:
        batch_label_action.append(data_dict["gt_action"][idx])
        batch_label_start.append(data_dict["gt_start"][idx])
        batch_label_end.append(data_dict["gt_end"][idx])
        batch_anchor_feature.append(data_dict["feature"][idx])

    batch_label_action=np.array(batch_label_action)
    batch_label_start=np.array(batch_label_start)
    batch_label_end=np.array(batch_label_end)
    batch_anchor_feature=np.array(batch_anchor_feature)
    batch_anchor_feature=np.reshape(batch_anchor_feature,[len(video_list),tscale,-1])
    return batch_label_action,batch_label_start,batch_label_end,batch_anchor_feature

def getFullData(dataSet):
    """Load full data in dataset
    """
    train_dict,val_dict,test_dict=getDatasetDict()
    if dataSet=="train":
        video_dict=train_dict
    else:
        video_dict=val_dict
    video_list=video_dict.keys()

    batch_bbox=[]
    batch_index=[0]
    batch_anchor_xmin=[]
    batch_anchor_xmax=[]
    batch_anchor_feature=[]
    for i in range(len(video_list)):
        if i%100==0:
            print "%d / %d %s videos are loaded" %(i,len(video_list),dataSet)
        video_name=video_list[i]
        video_info=video_dict[video_name]
        video_frame=video_info['duration_frame']
        video_second=video_info['duration_second']
        feature_frame=video_info['feature_frame']
        corrected_second=float(feature_frame)/video_frame*video_second
        video_labels=video_info['annotations']
        for j in range(len(video_labels)):
            tmp_info=video_labels[j]
            tmp_start=tmp_info['segment'][0]
            tmp_end=tmp_info['segment'][1]
            tmp_start=max(min(1,tmp_start/corrected_second),0)
            tmp_end=max(min(1,tmp_end/corrected_second),0)
            batch_bbox.append([tmp_start,tmp_end])

        tmp_anchor_xmin=[tgap*i for i in range(tscale)]
        tmp_anchor_xmax=[tgap*i for i in range(1,tscale+1)]
        batch_anchor_xmin.append(list(tmp_anchor_xmin))
        batch_anchor_xmax.append(list(tmp_anchor_xmax))
        batch_index.append(batch_index[-1]+len(video_labels))
        tmp_df=pd.read_csv("./data/activitynet_feature_cuhk/csv_mean_"+str(tscale)+"/"+video_name+".csv")
        batch_anchor_feature.append(tmp_df.values[:,:])
    num_data=len(batch_anchor_feature)
    batch_label_action=[]
    batch_label_start=[]
    batch_label_end=[]

    for idx in range(num_data):
        gt_bbox=np.array(batch_bbox[batch_index[idx]:batch_index[idx+1]])
        #break
        gt_xmins=gt_bbox[:,0]
        gt_xmaxs=gt_bbox[:,1]
        anchor_xmin=batch_anchor_xmin[idx]
        anchor_xmax=batch_anchor_xmax[idx]

        gt_lens=gt_xmaxs-gt_xmins
        gt_len_small=np.maximum(tgap,0.1*gt_lens)

        gt_start_bboxs=np.stack((gt_xmins-gt_len_small/2,gt_xmins+gt_len_small/2),axis=1)
        gt_end_bboxs=np.stack((gt_xmaxs-gt_len_small/2,gt_xmaxs+gt_len_small/2),axis=1)

        match_score_action=[]
        for jdx in range(len(anchor_xmin)):
            match_score_action.append(np.max(ioa_with_anchors(anchor_xmin[jdx],anchor_xmax[jdx],gt_xmins,gt_xmaxs)))
        match_score_start=[]
        for jdx in range(len(anchor_xmin)):
            match_score_start.append(np.max(ioa_with_anchors(anchor_xmin[jdx],anchor_xmax[jdx],gt_start_bboxs[:,0],gt_start_bboxs[:,1])))
        match_score_end=[]
        for jdx in range(len(anchor_xmin)):
            match_score_end.append(np.max(ioa_with_anchors(anchor_xmin[jdx],anchor_xmax[jdx],gt_end_bboxs[:,0],gt_end_bboxs[:,1])))

        batch_label_action.append(match_score_action)
        batch_label_start.append(match_score_start)
        batch_label_end.append(match_score_end)

    dataDict={"gt_action":batch_label_action,"gt_start":batch_label_start,"gt_end":batch_label_end,"feature":batch_anchor_feature}
    return dataDict

def getProposalDataTest(video_list,video_dict):
    """Load data during testing
    """
    batch_anchor_xmin=[]
    batch_anchor_xmax=[]
    batch_anchor_feature=[]
    for i in range(len(video_list)):
        video_name=video_list[i]
        tmp_anchor_xmin=[tgap*i for i in range(tscale)]
        tmp_anchor_xmax=[tgap*i for i in range(1,tscale+1)]
        batch_anchor_xmin.append(list(tmp_anchor_xmin))
        batch_anchor_xmax.append(list(tmp_anchor_xmax))
        tmp_df=pd.read_csv("./data/activitynet_feature_cuhk/csv_mean_"+str(tscale)+"/"+video_name+".csv")
        batch_anchor_feature.append(tmp_df.values[:,:])
    batch_anchor_xmin=np.array(batch_anchor_xmin)
    batch_anchor_xmax=np.array(batch_anchor_xmax)
    batch_anchor_feature=np.array(batch_anchor_feature)
    batch_anchor_feature=np.reshape(batch_anchor_feature,[len(video_list),tscale,-1])
    return batch_anchor_xmin,batch_anchor_xmax,batch_anchor_feature