bartzi / loans Goto Github PK
View Code? Open in Web Editor NEWCode for the Paper "LoANs: Weakly Supervised Object Detection with Localizer Assessor Networks"
License: GNU General Public License v3.0
Code for the Paper "LoANs: Weakly Supervised Object Detection with Localizer Assessor Networks"
License: GNU General Public License v3.0
Hi ,
I am trying to download figure_skating_assessor_datasets.zip file from ( https://bartzi.de/documents/attachment/download?hash_value=8a7b6af749bd5aa0d2c2d3ff0725ebca_24 .) .
Unfortunately I am not able to extract the file and when I check the file is only 8 Bytes in size .
Is there any other link where I can download the dataset ?
Hi,
What is actual training data for Assessor ?
I have observed that inside the images.csv file , ratios are generated only if I am using
--zoom-mode
For example,
In the created dataset (Templates are industry workers and Backgrounds are production plant)
in Figure B the ratio represents correctly
but in Figure A and C, the ratio do not seem to correct representation of the IoU between Template and Background image ( In figure C, 0.85 means 85% of the image is covered by target object. Is that correct ?)
Could you please help me to identify, what exactly is the ground truth for Assessor ?
Thanks
Rahul
I have followed the instructions given in the page and have tried to visualize the results using following code
### with video_sheeping.py
python video_sheeping.py figure_skating/models/resnet_50_augmentation_no_noise_75_100/Resnet50SheepLocalizer_97305.npz \log \-i videos/validation_data/video.mp4 \ -g 0 \ -o validation_data/analyzed_video.mp4
And I get the following error
0%| | 0/5 [00:00<?, ?it/s]
Could not find encoder for codec id 27: Encoder not found
Traceback (most recent call last): | 0/11985 [00:00<?, ?it/s]
File "video_sheeping.py", line 107, in
sheep(args, localizer, video)
File "video_sheeping.py", line 67, in sheep
render_scores=args.discriminator is not None
AttributeError: 'Namespace' object has no attribute 'discriminator'Exception ignored in: <bound method tqdm.del of 0%| | 0/11985 [00:17<?, ?it/s]>
Traceback (most recent call last):
File "/home/rahul/.virtualenvs/loans/lib/python3.6/site-packages/tqdm/_tqdm.py", line 931, in del
self.close()
File "/home/rahul/.virtualenvs/loans/lib/python3.6/site-packages/tqdm/_tqdm.py", line 1133, in close
self._decr_instances(self)
File "/home/rahul/.virtualenvs/loans/lib/python3.6/site-packages/tqdm/_tqdm.py", line 496, in _decr_instances
cls.monitor.exit()
File "/home/rahul/.virtualenvs/loans/lib/python3.6/site-packages/tqdm/_monitor.py", line 52, in exit
self.join()
File "/usr/lib/python3.6/threading.py", line 1053, in join
raise RuntimeError("cannot join current thread")
RuntimeError: cannot join current thread
command used as following
python image_sheeping.py figure_skating/models/resnet_50_augmentation_no_noise_75_100/Resnet50SheepLocalizer_97305.npz \log \-i figure_skating/evaluation_dataset/test_images/*.png \ -g 0 \ -o validation_data/images/analyzed
Output/error
0%| | 0/6 [00:00<?, ?it/s]Traceback (most recent call last):
File "image_sheeping.py", line 44, in
bboxes, scores = localizer.localize(processed_image)
ValueError: too many values to unpack (expected 2)
Exception ignored in: <bound method tqdm.del of 0%| | 0/6 [00:15<?, ?it/s]>
Traceback (most recent call last):
File "/home/rahul/.virtualenvs/loans/lib/python3.6/site-packages/tqdm/_tqdm.py", line 931, in del
self.close()
File "/home/rahul/.virtualenvs/loans/lib/python3.6/site-packages/tqdm/_tqdm.py", line 1133, in close
self._decr_instances(self)
File "/home/rahul/.virtualenvs/loans/lib/python3.6/site-packages/tqdm/_tqdm.py", line 496, in _decr_instances
cls.monitor.exit()
File "/home/rahul/.virtualenvs/loans/lib/python3.6/site-packages/tqdm/_monitor.py", line 52, in exit
self.join()
File "/usr/lib/python3.6/threading.py", line 1053, in join
raise RuntimeError("cannot join current thread")
RuntimeError: cannot join current thread
Can anyone suggest a possible solution ?
A detailed tutorial screen-capture video of how to use this for labeling/segmentation would help. I know it's a lot to ask for.
Hi ,
I read the paper and it is really an interesting work.
I have a doubt regarding it. Will the network be able to predict bounding box on a new image (lets say a random image of figure skater downloaded from internet ) ?
If it is possible , is there any test script available which can use trained weights on your figure-skating dataset ?
Also can you share the details of how the input data to assessor and localizer looks like ?
Thanks
Rahul
Hi ,
I have created the dataset as per the readme and when I run train_sheep_localizer.py
as in following command :
python /home/rahul/ActiveShuttle/WSOL/LOANS/loans-master/train_sheep_localizer.py /home/rahul/ActiveShuttle/WSOL/LOANS/Dataset/Localizer/pics_extracted/gt.csv /home/rahul/ActiveShuttle/WSOL/LOANS/figure_skating/evaluation_dataset/gt.json /home/rahul/ActiveShuttle/WSOL/LOANS/Dataset/Assessor/dataset/images.csv --target-size 75 100 --batch-size 64 --gpu 0 --learning-rate 1e4 --log-name figure_skating_person --use-resnet-18
I get following error
Traceback (most recent call last):
File "/home/rahul/ActiveShuttle/WSOL/LOANS/loans-master/train_sheep_localizer.py", line 259, in <module>
main()
File "/home/rahul/ActiveShuttle/WSOL/LOANS/loans-master/train_sheep_localizer.py", line 155, in main
**updater_args
File "/home/rahul/ActiveShuttle/WSOL/LOANS/loans-master/sheep/sheep_updater.py", line 12, in __init__
self.n_dis = kwargs.pop('n_dis')
KeyError: 'n_dis'
Can anyone please help to find the problem ?
@Bartzi
Thanks for providing the code. Here is a minor typo in the readme:
s/you object/your object
s/on wich/on which
s/the scipt/the script
s/fileholding/file holding
I would look over the entire readme, and maybe have a collaborator look over it as well.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.