Comments (8)
Dear @Mshz2 ,
I noticed that your ground-truth files (XML) name the objects class as "pore" (<name>pore</name>
), and your detection files identify the object classes as "0" (all your detection files start with "0"). So, the problem is because the "0" could either be the id
or the name
of the class.
First thing you have to do is: separate your files into 3 different folders: groundtruths (XML files), detections (.txt files) and images.
Then, please, update your code with the fixes I have just made. The error messages were not telling you the error correctly. I fixed it.
Now you have two choices. The first one is the easiest one :)
1st option:
Make a txt file as this one
class.txt
As you can see, in its first line (line 0) I put the name of your object ("spore"). This way the program will know that your detection files are representing the class "0" with the name in the line 0 of the file ("spore").
Set the interface with the following options:
Ground truth:
Annotations: Choose the directory where your (.xml) files are.
Images (optional): Choose the directory where your images are.
Coordinate type: Pick the option "PASCAL VOC (.xml)
Output: Choose a directory where the outputs will be saved.
Detections:
Annotations: Choose the directory where your detections (.txt) files are.
Classes: Point to the class.txt
file.
Coordinate type: Pick the option <class_id> <confidence> <left> <top> <right> <bottom> (ABSOLUTE)
Output: Choose a directory where the outputs will be saved.
2nd option:
Change all detection (.txt) files replacing the first element "0" by "pore".
Example:
Now you have: 0 0.99 37 143 98 206
Change it to: pore 0.99 37 143 98 206
Set the interface with the following options:
Ground truth:
Annotations: Choose the directory where your (.xml) files are.
Images: (optional) Choose the directory where your images are.
Coordinate type: Pick the option "PASCAL VOC (.xml)
Detections:
Annotations: Choose the directory where your (.txt) files are.
Classes: Leave it blank, because you now included the name of the classes in your detection files.
Coordinate type: Pick the option <class_name> <confidence> <left> <top> <right> <bottom> (ABSOLUTE)
Output: Choose a directory where the outputs will be saved.
In both cases, if you inserted the directory with the images, you can visualize your bounding boxes in the statistics.
I hope it helps.
Best regards
from review_object_detection_metrics.
@Mshz2
This is strange. With the files you provided, I obtain different results:
Have you updated the repository? I updated it with few changes.
You dont have to show them here, but if you click on "show detections statistics", are you able to see the images with the bounding boxes?
from review_object_detection_metrics.
Great! :)
I will close this issue now. If you have any other problem, just open another issue or reach me at linkedin.
Best regards.
from review_object_detection_metrics.
Hi @Mshz2 ,
Could you please, zip a couple images, detections and ground-truths and attach here, so I can try to replicate this problem?
Thank you
from review_object_detection_metrics.
Hi @Mshz2 ,
Could you please, zip a couple images, detections and ground-truths and attach here, so I can try to replicate this problem?
Thank you
Thanks a lot.
test.zip
btw sorry, I had to delete images and give just a blank white frame as I am not allowed to share images. But thanks for the support, I hope this would be usefull.
from review_object_detection_metrics.
No problem you had to delete the images. I totally understand it. :)
I hope the instructions in the previous message helps you. Let me know if you got it working.
from review_object_detection_metrics.
@rafaelpadilla thanks aloooot for your effort.
I did as you said but unfortunately I get everything 0. :(
from review_object_detection_metrics.
@rafaelpadilla forget above message :) seems like when I did as second option it worked!
In first option it gave everything 0
from review_object_detection_metrics.
Related Issues (20)
- Equation (9) and (11) in publication HOT 3
- how to use commond line ? HOT 1
- Do you have the detection results of different detection models in coco json format. HOT 1
- could't open detected the annotation file made by yolov4 darknet HOT 2
- How to calculate the general Precision x recall curve of the detection model? HOT 1
- Coordinates format has repeated option HOT 4
- Can you share commands on how to run the coco and PascalVOC evaluators via terminal to generate the outputs? HOT 3
- No results HOT 4
- Feature request: Docker containerization HOT 3
- move(self, int, int): argument 1 has unexpected type 'float' HOT 5
- ModuleNotFoundError: No module named 'src' HOT 4
- Bug in the converter.py while looking for images folder ? HOT 1
- AttributeError: 'list' object has no attribute 'items' HOT 1
- Please provide command line to evaluate on terminal HOT 1
- Issue during start: TypeError: arguments did not match any overloaded call HOT 3
- np.bool is deprecated HOT 5
- numpy: set_window_title moved to manager HOT 3
- Building the conda environment causes an endless loop HOT 2
- Result is 0 HOT 3
- Getting precision, recall, F1 values HOT 2
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 review_object_detection_metrics.