Comments (12)
Hi, below are my answers:
Q: "there will be some frames that are not part of the action"
A: There is a file named "valid_images.txt" inside the dataset. It specifies the label of each video clip. So all frames should be part of an action.
I paste part of the file at below so you can check it.
Q: "can I replace the data with time series relationship with the collected pictures"
A: I don't know your project, but I guess you can -- You can try and play with different algorithms that you come up with.
Contents of "valid_images.txt":
jump_03-02-12-34-01-795
52 59
72 79
114 121
134 141
161 169
181 188
236 245
276 284
313 321
331 338
349 356
367 374
385 392
403 409
424 428
441 541
jump_03-12-09-18-26-176
58 680
jump_03-13-11-27-50-720
65 393
kick_03-02-12-36-05-185
54 62
75 84
102 108
128 136
148 156
167 174
194 202
215 223
237 245
260 267
281 288
302 310
330 336
348 355
366 374
391 398
409 416
428 435
kick_03-08-20-32-41-586
49 110
143 196
kick_03-12-09-23-41-176
58 755
kick_03-13-16-18-12-361
71 150
203 237
264 303
309 349
punch_03-12-09-21-27-876
60 858
run_03-02-12-31-47-095
62 403
run_03-12-09-15-25-375
70 760
sit_03-02-12-27-51-085
65 217
sit_03-02-12-28-32-893
74 355
sit_03-12-09-25-43-527
52 836
sit_03-13-16-15-56-861
55 70
90 103
124 156
178 297
318 363
383 413
433 553
575 616
638 673
692 726
746 777
797 827
893 961
980 1005
1024 1059
squat_03-08-20-26-57-195
75 785
squat_03-13-13-21-48-761
41 293
stand_03-08-20-24-55-587
55 520
stand_03-08-20-35-06-287
60 500
stand_03-12-09-17-05-376
35 395
stand_03-12-09-17-05-376
481 619
stand_03-13-13-22-37-869
40 335
walk_03-12-09-13-10-875
60 800
walk_03-02-12-30-23-393
124 602
wave_03-08-20-28-39-387
65 995
wave_03-13-13-23-25-262
47 354
from realtime-action-recognition.
Thank you for your answer, I have seen the video segmentation you gave, I don’t know if there is a standard for this interval, such as at least how many frames, or other quantity requirements, thank you very much.
from realtime-action-recognition.
No standard. I thought this was correct, so I labeled it in this way.
from realtime-action-recognition.
Hello, I have preprocessed with the data you provided, and the following problem has occurred:
File "src/scripts/skeletons_info_generator.py", line 40, in
skeleton = skeletons[idx_person]
IndexError: list index out of range
I don't know what caused it. Do you have such a problem in the process of processing? @felixchenfy thank you
from realtime-action-recognition.
No problem on my laptop.
Since idx_person=0, the error "IndexError: list index out of range" indicates that the list is empty. Maybe there is some problem with path settings.
This script simply reads a bunch of .txt files, and put their contents into another .txt file. I believe you can easily debug it ....
from realtime-action-recognition.
hello,are you Chinese?@felixchenfy
from realtime-action-recognition.
恩
from realtime-action-recognition.
你好,不知道可以加一下你其他的联系方式请教一些问题吗?非常感谢了,[email protected] @felixchenfy
from realtime-action-recognition.
已加qq
from realtime-action-recognition.
Hello
what does the number behind the action mean ?
for example sit_03-12-09-25-43-527, what does 03-12-09-25-43-527 those number mean ?
thanks
from realtime-action-recognition.
Hello
what does the number behind the action mean ?
for example sit_03-12-09-25-43-527, what does 03-12-09-25-43-527 those number mean ?thanks
@huwade Hi, It's the time that I recorded the images: 2019 March 12th, 9 am, 25 minutes, 43 seconds, 527 milliseconds.
from realtime-action-recognition.
thanks
from realtime-action-recognition.
Related Issues (20)
- Increase number of person dectected HOT 1
- How to use [lying, other, sitting, standing]
- ModuleNotFoundError: No module named 'sklearn.neighbors.classification' HOT 8
- 大佬您好,想问一下您的数据集怎么标定的吗?如何生成valid_images.txt文件呢 HOT 1
- 行为识别用的什么方法呀 HOT 1
- training is slow !! HOT 2
- how to remove the head joints HOT 1
- ModuleNotFoundError: No module named 'tf_pose' HOT 3
- 您好!可以请教一下是否有相关文章或者资料对这个项目的介绍分析吗 HOT 7
- 不好意思,请问在运作前都会显示这些东西而延迟一段时间,请问这个要怎去掉呢?
- Normalization performed twice
- Is it possible to use LSTM models using this feature preprocessing and extraction method?
- number of features
- AttributeError: module 'tensorflow_core.contrib' has no attribute 'tensorrt'
- how to generate valid_images.txt. HOT 2
- 当我测试摄像头的时候,CV2
- warnings.warn("Invalid fps of video: {}".format(video_path))
- Converting to RNN HOT 1
- AttributeError: module 'tensorflow' has no attribute 'nn' HOT 1
- can i use ucf101dataset to train more labels?
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from realtime-action-recognition.