-
The iSAID paper uses 1000 instead of the default 100.
-
I am using the default configuration file and getting multiple detections on the same object. NMS is not working properly. Can you tell me why ?
-
### Search before asking
- [X] I have searched the Ultralytics YOLO [issues](https://github.com/ultralytics/ultralytics/issues) and [discussions](https://github.com/ultralytics/ultralytics/discussion…
-
Player boxes should stay in order, maybe bounce off each other?
-
### Search before asking
- [X] I have searched the HUB [issues](https://github.com/ultralytics/hub/issues) and [discussions](https://github.com/ultralytics/hub/discussions) and found no similar quest…
-
This Processing took 37 ms. My question is whether this running time of detection normal. And another problem is that the minimum distance data of the two detected objects shows abnormality. I don't k…
-
When I used your source code to transform the yolov4 model I had trained, the TF weight file I got was tested. The detection box shown in the test results was too small to completely cover the whole t…
-
If no detections are found, you get an error on the `nms_keep` variable in model.refine_detections(). Defining nms_keep before this loop as an empty tensor bring up errors elsewhere.
UnboundLocalEr…
-
Hi,
I am using the detectron2 cascade mask rcnn vitdet h checkpoint in the demo for hamer and it is very slow for inference on one image compared to detectron2 model zoo numbers. I get .78s/im vs the…
-
First, thanks for making this. It's super useful!
## The Problem
For objects that are very small compared to the total image size, the label might block the object.
I'm currently converting thi…