forked from AIVIETNAMResearch/Video-Text-Retrieval
-
Notifications
You must be signed in to change notification settings - Fork 0
/
app.py
639 lines (500 loc) · 24.4 KB
/
app.py
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
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
from flask import Flask, render_template, Response, request, send_file, jsonify
import cv2
import os
import torch
import sys
import pandas as pd
import json
import faiss
import numpy as np
os.environ['CUDA_VISIBLE_DEVICES'] = '-1'
# from utils.nlp_processing import Translation
from utils.translate_vi2en import translate_vi2en
from pathlib import Path
from posixpath import join
from langdetect import detect
from utils.faiss_processing import write_csv, extract_feats_from_bin, save_feats_to_bin, get_id2index, \
load_json_file,load_bin_file,mapping_index, search_tags, get_all_ids, remove_keys_and_save, search_image2image, \
searchcontinues, read_index_file, save_bin_delete_noise
from utils.submit import write_csv, show_csv
# from sentence_transformers import SentenceTransformer, util
# from utils.ocr_processing import fill_ocr_results, fill_ocr_df
from utils.group_keyframes import convertArray
current_dir = os.path.dirname(os.path.abspath(__file__))
# Xác định đường dẫn tới thư mục LAVIS
lavis_dir = os.path.join(current_dir, 'LAVIS')
# Thêm đường dẫn tương đối của thư mục LAVIS vào sys.path
sys.path.append(lavis_dir)
from lavis.models import load_model_and_preprocess
# http://0.0.0.0:5001/thumbnailimg?index=0
# app = Flask(__name__, template_folder='templates', static_folder='static')
app = Flask(__name__, template_folder='templates')
# Faiss
bin_file='dict/faiss_blip_v1_cosine_final.bin'
json_path = 'dict/dict_final/keyframes_id.json'
json_id2img_path = 'dict/dict_final/dict_image_path_id2img.json'
json_img2id_path = 'dict/dict_final/dict_image_path_img2id.json'
json_keyframe2id = 'dict/dict_final/keyframe_path2id.json'
json_keyframe2path = 'dict/dict_final/keyframe_id2path.json'
file_path = 'search_continues/list_index_search_continues.txt'
################# LOAD FILE BIN ##################
faiss_model = load_bin_file(bin_file)
############## MODEL BLIP #################
__device = "cuda" if torch.cuda.is_available() else "cpu"
model, vis_processors_blip, text_processors_blip = load_model_and_preprocess("blip_image_text_matching",
"base",
device=__device,
is_eval=True)
# with open("dict/info_ocr.txt", "r", encoding="utf8") as fi:
# ListOcrResults = list(map(lambda x: x.replace("\n",""), fi.readlines()))
# with open("dict/info_asr.txt", "r", encoding="utf8") as fi:
# ListASRResults = list(map(lambda x: x.replace("\n",""), fi.readlines()))
# df_asr = pd.read_csv("dict/info_asr.txt", delimiter=",", header=None)
# df_asr.columns = ["video_id", "frame_id", "asr"]
with open(json_id2img_path, 'r') as f:
DictId2Img = json.loads(f.read())
with open(json_img2id_path, 'r') as f:
DictImg2Id = json.loads(f.read())
with open(json_keyframe2path, 'r') as f:
DictKeyframe2Path = json.loads(f.read())
with open(json_keyframe2id, 'r') as f:
DictKeyframe2Id = json.loads(f.read())
LenDictPath = len(load_json_file(json_path))
DictImagePath = load_json_file(json_path)
######################### HOME PAGE ########################################
@app.route('/thumbnailimg')
def thumbnailimg():
print("load_iddoc")
# remove old file submit
submit_path = join("submission", "submit.csv")
old_submit_path = Path(submit_path)
if old_submit_path.is_file():
os.remove(submit_path)
# open(submit_path, 'w').close()
temp_faiss_path = join("search_continues", "temp_faiss.bin")
old_temp_path = Path(temp_faiss_path)
if old_temp_path.is_file():
os.remove(old_temp_path)
temp_txt_path = join("search_continues", "list_index_search_continues.txt")
old_temp_path = Path(temp_txt_path)
if old_temp_path.is_file():
os.remove(old_temp_path)
objs_path = Path('search_continues/objs.csv')
if objs_path.is_file():
os.remove(objs_path)
if os.path.exists('search_continues/after_rm_noise.bin'):
os.remove('search_continues/after_rm_noise.bin')
if os.path.exists('search_continues/after_rm_noise_idx.txt'):
os.remove('search_continues/after_rm_noise_idx.txt')
if os.path.exists('search_continues/after_dict_rm_noise.json'):
os.remove('search_continues/after_dict_rm_noise.json')
# bin_file = 'dict/faiss_blip_v1_cosine_final.bin'
print("LenDictPath: ", LenDictPath)
pagefile = []
index = int(request.args.get('index'))
if index == None:
index = 0
imgperindex = 100
pagefile = []
page_filelist = []
list_idx = []
# print(index)
if LenDictPath-1 > index+imgperindex:
first_index = index * imgperindex
last_index = index*imgperindex + imgperindex
tmp_index = first_index
while tmp_index < last_index:
page_filelist.append(DictImagePath[tmp_index]["image_path"])
list_idx.append(tmp_index)
tmp_index += 1
else:
first_index = index * imgperindex
last_index = LenDictPath
tmp_index = first_index
while tmp_index < last_index:
page_filelist.append(DictImagePath[tmp_index]["image_path"])
list_idx.append(tmp_index)
tmp_index += 1
for imgpath, id in zip(page_filelist, list_idx):
pagefile.append({'imgpath': imgpath, 'id': id})
pagefile_new = convertArray(pagefile)
data = {'num_page': int(LenDictPath/imgperindex)+1, 'pagefile': pagefile_new}
return render_template('home.html', data=data)
###################### SEARCH IMAGE PATH #####################
@app.route('/search_image_path')
def search_image_path():
pagefile = []
frame_path = request.args.get('frame_path')
list_frame_split = frame_path.split("/")
video_dir = list_frame_split[0]
image_name = list_frame_split[1] + ".jpg"
keyframe_dir = video_dir.split('_')[0]
frame_path = join("Database", "Keyframes_"+keyframe_dir, video_dir, image_name)
frame_path = frame_path.replace("\\","/")
frame_id = DictKeyframe2Id[frame_path]
imgperindex = 100
pagefile.append({'imgpath': frame_path, 'id': int(frame_id)})
# show around 40 key image
total_image_in_video = int(DictImg2Id[keyframe_dir][video_dir]["total_image"])
number_image_id_in_video = int(DictImg2Id[keyframe_dir][video_dir][image_name])
first_index_in_video = number_image_id_in_video-50 if number_image_id_in_video-50>0 else 0
last_index_in_video = number_image_id_in_video+50 if number_image_id_in_video+50<total_image_in_video else total_image_in_video
frame_index = first_index_in_video
while frame_index < last_index_in_video:
new_frame_name = DictId2Img[keyframe_dir][video_dir][str(frame_index)]
frame_in_video_path = join("Database", "Keyframes_"+keyframe_dir, video_dir, new_frame_name)
frame_in_video_path = frame_in_video_path.replace("\\","/")
if frame_in_video_path in DictKeyframe2Id:
frame_id_in_video_path = DictKeyframe2Id[frame_in_video_path]
pagefile.append({'imgpath': frame_in_video_path, 'id': int(frame_id_in_video_path)})
frame_index += 1
pagefile_new = convertArray(pagefile)
data = {'num_page': int(LenDictPath/imgperindex)+1, 'pagefile': pagefile_new}
return render_template('home.html', data=data)
####################### IMAGE SEARCH - SEARCH TO KEYFRAME ID ######################
import json
@app.route('/imgsearch')
def image_search():
print("image search")
pagefile = []
id_query = int(request.args.get('imgid'))
print('imgid: ', id_query)
k = request.args.get('topk')
k = int(k[3:])
# Check search continues
after_rm_noise_path = Path('search_continues/after_rm_noise.bin')
faiss_path = Path('search_continues/temp_faiss.bin')
if os.path.exists('search_continues/after_rm_noise_idx.txt'):
index_file = 'search_continues/after_rm_noise_idx.txt'
save_bin_delete_noise(bin_file, index_file, 'search_continues/after_rm_noise.bin')
if after_rm_noise_path.is_file(): # Mức độ ưu tiên của file bin search: after_rm_noise --> temp_faiss --> file bin gốc
print("continue searchingggg after delete noiseeee....................................")
new_file_bin = 'search_continues/after_rm_noise.bin'
index_file = 'search_continues/after_rm_noise_idx.txt'
scores, idx_images = searchcontinues(new_file_bin, index_file, k, id_query=id_query)
elif faiss_path.is_file():
print("continue searchinggg...........................................................")
new_file_bin = 'search_continues/temp_faiss.bin'
index_file = "search_continues/list_index_search_continues.txt"
scores, idx_images = searchcontinues(new_file_bin, index_file, k, id_query=id_query)
else:
query_feats = faiss_model.reconstruct(id_query).reshape(1,-1)
scores, idx_images = faiss_model.search(query_feats, k=k)
idx_images = idx_images.flatten()
scores = scores.flatten()
id2img_fps = DictImagePath
infos_query = list(map(id2img_fps.get, list(idx_images)))
image_paths = [info['image_path'] for info in infos_query]
scores = np.array(scores, dtype=np.float32).tolist()
imgperindex = 100
for imgpath, id, score in zip(image_paths, idx_images, scores):
pagefile.append({'imgpath': imgpath, 'id': int(id), 'score':score})
pagefile_new = convertArray(pagefile)
data = {'num_page': int(LenDictPath/imgperindex)+1, 'pagefile': pagefile_new}
return render_template('index_thumb1.html', data=data)
######################### SEARCH TO TEXT - BLIP SEARCH ############################
@app.route('/textsearch')
def text_search():
print("text search")
k = str(request.args.get('topk'))
k = int(k[3:])
pagefile = []
text_query = request.args.get('textquery')
if detect(text_query) == 'vi':
# translater = Translation()
# text = translater(text_query)
text = translate_vi2en(text_query)
else:
text = text_query
###### TEXT FEATURES EXACTING ######
txt = text_processors_blip["eval"](text)
text_features = model.encode_text(txt, __device).cpu().detach().numpy()
##### CHECK SEARCH CONTINUES #####
after_rm_noise_path = Path('search_continues/after_rm_noise.bin')
faiss_path = Path('search_continues/temp_faiss.bin')
###### SEARCHING #####
if os.path.exists('search_continues/after_rm_noise_idx.txt'):
index_file = 'search_continues/after_rm_noise_idx.txt'
save_bin_delete_noise(bin_file, index_file, 'search_continues/after_rm_noise.bin')
if after_rm_noise_path.is_file(): # Mức độ ưu tiên của file bin search: after_rm_noise --> temp_faiss --> file bin gốc
print("continue searchingggg after delete noiseeee....................................")
new_file_bin = 'search_continues/after_rm_noise.bin'
index_file = 'search_continues/after_rm_noise_idx.txt'
scores, idx_images = searchcontinues(new_file_bin, index_file, k, text_features=text_features)
elif faiss_path.is_file():
print("continue searchinggg...........................................................")
new_file_bin = 'search_continues/temp_faiss.bin'
index_file = "search_continues/list_index_search_continues.txt"
scores, idx_images = searchcontinues(new_file_bin, index_file, k, text_features=text_features)
else:
scores, idx_images = faiss_model.search(text_features, k=k)
idx_images = idx_images.flatten()
scores = scores.flatten()
###### GET INFOS KEYFRAMES_ID ######
id2img_fps = DictImagePath
infos_query = list(map(id2img_fps.get, list(idx_images)))
image_paths = [info['image_path'] for info in infos_query]
imgperindex = 100
scores = np.array(scores, dtype=np.float32).tolist()
# print(scores)
for imgpath, id, score in zip(image_paths, idx_images, scores):
pagefile.append({'imgpath': imgpath, 'id': int(id), 'score':score})
pagefile_new = convertArray(pagefile)
# print(pagefile_new)
data = {'num_page': int(LenDictPath/imgperindex)+1, 'pagefile': pagefile_new}
return render_template('index_thumb1.html', data=data)
####################### CONTINUES SEARCHINGGGG ####################
@app.route('/searchcontinues', methods=['POST'])
def search_continues():
data = request.get_json()
pagefile = data['pagelist']
# Sử dụng hàm để lấy danh sách tất cả các ID
indexs = get_all_ids(pagefile)
print(indexs)
new_bin_file = './search_continues/temp_faiss.bin'
new_list_idx_for_bin = './search_continues/list_index_search_continues.txt'
# bin_file = 'dict/faiss_blip_v1_cosine_final.bin'
if Path(new_list_idx_for_bin).is_file():
data_array = read_index_file(new_list_idx_for_bin)
ids = get_id2index(data_array) ## get index for index_list
print(ids)
feats = extract_feats_from_bin(bin_file, ids)
else:
feats = extract_feats_from_bin(bin_file, indexs)
# savefile sub bin and idx of frames
save_feats_to_bin(indexs, feats, new_bin_file, new_list_idx_for_bin)
after_rm_noise_path = Path('search_continues/after_rm_noise.bin')
after_rm_noise_txt_path = Path('search_continues/after_rm_noise_idx.txt')
if after_rm_noise_txt_path.is_file():
os.remove(after_rm_noise_txt_path)
if after_rm_noise_path.is_file():
os.remove(after_rm_noise_path)
print('Saved new bin file')
######################### GET FRAMES NEIGHTBOR ##############################
@app.route('/neighborsearch')
def neightbor_search():
print('neightbor frame search')
pagefile = []
id_query = int(request.args.get('imgid'))
list_shot_path = DictImagePath[id_query]['list_shot_path']
imgperindex = 100
for shot_info in list_shot_path:
pagefile.append({'imgpath': shot_info['shot_path'], 'id': int(DictKeyframe2Id[shot_info['shot_path']])})
# show around 200 key image
frame_path = DictImagePath[id_query]["image_path"]
video_dir = frame_path.split("/")[-2]
keyframe_dir = video_dir.split('_')[0]
image_name = frame_path.split("/")[-1]
total_image_in_video = int(DictImg2Id[keyframe_dir][video_dir]["total_image"])
number_image_id_in_video = int(DictImg2Id[keyframe_dir][video_dir][image_name])
first_index_in_video = number_image_id_in_video-50 if number_image_id_in_video-50>0 else 0
last_index_in_video = number_image_id_in_video+50 if number_image_id_in_video+50<total_image_in_video else total_image_in_video
frame_index = first_index_in_video
while frame_index < last_index_in_video:
new_frame_name = DictId2Img[keyframe_dir][video_dir][str(frame_index)]
frame_in_video_path = join("Database", "Keyframes_"+keyframe_dir, video_dir, new_frame_name)
frame_in_video_path = frame_in_video_path.replace("\\","/")
if frame_in_video_path in DictKeyframe2Id:
frame_id_in_video_path = DictKeyframe2Id[frame_in_video_path]
pagefile.append({'imgpath': frame_in_video_path, 'id': int(frame_id_in_video_path)})
frame_index += 1
pagefile_new = convertArray(pagefile)
# print(pagefile_new)
data = {'num_page': int(LenDictPath/imgperindex)+1, 'pagefile': pagefile_new}
return render_template('home.html', data=data)
####################### SEARCH FOR TAGS - SEARCH OBJ #####################
@app.route('/search_for_tags')
def search_for_tags():
print("search for tags...")
k = str(request.args.get('topk'))
k = int(k[3:])
pagefile = []
text_query = request.args.get('text_for_tags')
if Path('search_continues/objs.csv').is_file():
csv_file = 'search_continues/objs.csv'
else:
csv_file = 'dict/object_final.csv'
text_query = str(text_query)
print(text_query)
objs, idx_images, image_paths = search_tags(csv_file, text_query)
objs.to_csv('search_continues/objs.csv', index=False) # index=False để không lưu cột index
imgperindex = 100
for imgpath, id in zip(image_paths, idx_images):
pagefile.append({'imgpath': imgpath, 'id': int(id)})
pagefile_new = convertArray(pagefile)
data = {'num_page': int(LenDictPath/imgperindex)+1, 'pagefile': pagefile_new}
return render_template('index_thumb1.html', data=data)
########################## DELETE VIDEO ID NOISE #########################
@app.route('/delete_noise', methods=['POST'])
def delete_noise():
# Nhận dữ liệu từ yêu cầu POST
video_id = request.form.get('video_id')
print(video_id)
list_frame_json = request.form.get('list_frame')
# Tiến hành xóa nhiễu hoặc thực hiện các tác vụ khác ở đây
# new_bin_path = 'search_continues/after_rm_noise.bin'
new_list_idx_path = 'search_continues/after_rm_noise_idx.txt'
new_dict_path = 'search_continues/after_dict_rm_noise.json'
if os.path.exists(new_dict_path):
ids, new_dict = remove_keys_and_save(video_id, new_dict_path)
else:
ids, new_dict = remove_keys_and_save(video_id, json_keyframe2id)
with open(new_list_idx_path, 'w') as output_file:
json.dump(ids, output_file)
with open(new_dict_path, 'w') as output:
json.dump(new_dict, output)
# bin_file = 'dict/faiss_blip_v1_cosine_final.bin'
# ids, feats = extract_feats_from_bin(bin_file, ids)
# # savefile sub bin and idx of frames
# index = faiss.IndexFlatIP(256)
# index.add(feats)
# faiss.write_index(index, new_bin_path)
# print('Saved new bin file')
# Trả về phản hồi (nếu cần)
return 'Success: Saved delete noiseee!!!', 200
########################## WRITE CSV #########################################
@app.route('/writecsv')
def submit():
print("writecsv")
info_key = request.args.get('info_key')
mode_write_csv = request.args.get('mode')
print("info_key", info_key)
print("mode: ", mode_write_csv)
info_key = info_key.split(",")
id_query = int(info_key[0])
selected_image = info_key[1]
number_line, list_frame_id = write_csv(DictImagePath, mode_write_csv, selected_image, id_query, "submission")
str_fname = ",".join(list_frame_id[:])
# str_fname += " #### number csv line: {}".format(number_line)
info = {
"str_fname": str_fname,
"number_line": str(number_line)
}
return jsonify(info)
################# GET IMAGES FOR DISPLAY #################
@app.route('/get_img')
def get_img():
# print("get_img")
fpath = request.args.get('fpath')
# fpath = fpath
list_image_name = fpath.split("/")
# image_name = "/".join(list_image_name[-2:])
image_name = list_image_name[-1].split('.')[0]
if os.path.exists(fpath):
img = cv2.imread(fpath)
else:
print("load 404.jph")
img = cv2.imread("./static/images/404.jpg")
img = cv2.resize(img, (1280, 720))
# Tọa độ và kích thước hình chữ nhật nền
x, y = 0, 0 # Tọa độ góc trái trên cùng của hình chữ nhật
w, h = cv2.getTextSize(image_name, cv2.FONT_HERSHEY_SIMPLEX, 3, 6)[0]
padding = 10 # Khoảng cách giữa văn bản và hình chữ nhật
# Vẽ hình chữ nhật nền
cv2.rectangle(img, (x, y), (x + w + padding, y + h + padding), (217, 217, 217), -1) # Màu nền #D9D9D9
# Vẽ văn bản
cv2.putText(img, image_name, (x + padding, y + h + padding), cv2.FONT_HERSHEY_SIMPLEX, 3, (0, 0, 0), 5, cv2.LINE_AA) # Màu chữ đen
ret, jpeg = cv2.imencode('.jpg', img)
return Response((b'--frame\r\n'
b'Content-Type: image/jpeg\r\n\r\n' + jpeg.tobytes() + b'\r\n\r\n'),
mimetype='multipart/x-mixed-replace; boundary=frame')
########################## DOWLOAD SUBMIT FILE #########################
@app.route('/dowload_submit_file', methods=['GET'])
def dowload_submit_file():
print("dowload_submit_file")
filename = request.args.get('filename')
fpath = join("submission", filename)
print("fpath", fpath)
return send_file(fpath, as_attachment=True)
########################## GET FIRST ROW #########################
@app.route('/get_first_row')
def getFirstRowOfCsv():
csv_path = "submission/submit.csv"
result = {
'video_id':"None",
'frame_id':"None"
}
if os.path.exists(csv_path):
lst_frame = show_csv(csv_path)[0]
print(lst_frame)
video_id, frame_id = lst_frame.split("/")[-2:]
result["video_id"] = video_id
result["frame_id"] = int(frame_id[:-4])
return result
################# VISUALIZE FRAME SELECTED #################################
@app.route('/visualize')
def visualize():
number_of_query = int(request.args.get('number_of_query'))
csv_path = join("submission", "query-{}.csv".format(number_of_query))
query_path = join("query","query-{}.txt".format(number_of_query))
if os.path.exists(query_path):
with open(query_path, "rb") as fi:
query_content = fi.read().decode("utf-8").replace(" ","_")
pagefile = []
lst_frame = show_csv(csv_path)
for frame_path in lst_frame:
frame_id = DictKeyframe2Id[frame_path]
pagefile.append({'imgpath': frame_path, 'id': int(frame_id)})
pagefile_new = convertArray(pagefile)
if query_content is not None:
data = {'num_page': 1, 'pagefile': pagefile_new, 'query': query_content}
else:
data = {'num_page': 1, 'pagefile': pagefile_new}
return render_template('index_thumb1.html', data=data)
# @app.route('/ocrfilter')
# def ocrfilter():
# print("ocr search")
# pagefile = []
# text_query = request.args.get('text_ocr')
# list_all = fill_ocr_results(text_query, ListOcrResults)
# list_all.extend(fill_ocr_results(text_query, ListASRResults))
# # list_all = fill_ocr_df(text_query, df_ocr)
# # list_all = np.vstack((list_all, fill_ocr_df(text_query, df_ocr)))
# print("list results of ocr + asr: ", list_all)
# imgperindex = 100
# for frame in list_all:
# list_frame_name = frame.split("/")
# keyframe_dir = list_frame_name[0][:7]
# video_dir = list_frame_name[0]
# new_frame_name = list_frame_name[-1]
# frame_in_video_path = join("Database", "KeyFrames"+keyframe_dir, video_dir, new_frame_name)
# frame_in_video_path = frame_in_video_path.replace("\\","/")
# # print("frame_in_video_path: ", frame_in_video_path)
# if frame_in_video_path in DictKeyframe2Id:
# print("frame_in_video_path: ", frame_in_video_path)
# frame_id_in_video_path = DictKeyframe2Id[frame_in_video_path]
# pagefile.append({'imgpath': frame_in_video_path, 'id': int(frame_id_in_video_path)})
# data = {'num_page': int(LenDictPath/imgperindex)+1, 'pagefile': pagefile}
# return render_template('index_thumb.html', data=data)
from PIL import Image
@app.route('/img2imgs', methods=['GET', 'POST'])
def index():
directory_path = "search_continues/uploaded/"
if request.method == 'POST':
file = request.files['query_img']
# Save query image
img = Image.open(file.stream) # PIL image
if not os.path.exists(directory_path):
os.makedirs(directory_path)
uploaded_img_path = directory_path + file.filename
img.save(uploaded_img_path)
idx_image, scores = search_image2image(uploaded_img_path, )
# Lấy kết quả và gửi đến html
id2img_fps = DictImagePath
infos_query = list(map(id2img_fps.get, list(idx_image)))
image_paths = [info['image_path'] for info in infos_query]
scores = np.array(scores, dtype=np.float32).tolist()
print("searching.......")
imgperindex = 100
pagefile = []
for imgpath, id, score in zip(image_paths, idx_image, scores):
pagefile.append({'imgpath': imgpath, 'id': int(id), 'score':score})
print("searching.........")
pagefile_new = convertArray(pagefile)
data = {'num_page': int(LenDictPath/imgperindex)+1, 'pagefile': pagefile_new}
return render_template('index_thumb1.html', data=data)
if __name__ == '__main__':
submit_dir = "submission"
if not os.path.exists(submit_dir):
os.mkdir(submit_dir)
app.run(debug=False, host="0.0.0.0", port=5001)