-
Should this two losses be added to keras_cv.losses ?
1. DIoU (Distance over IoU) Loss is independent to the size of the object being detected. It is seen to converge mush faster than GIoU even with…
-
1、Unfair comparison between state-of-the-art methods. State-of-the-art detectors such as VFNet don't use any data augmentations and only train for 24 epochs, but ObjectBox uses the CutMix and Mosaic …
-
[YOLOv2](https://arxiv.org/pdf/1612.08242.pdf) states that when using kmeans they use a custom distance function to get the anchors because using standard kmeans "with Euclidean distance, larger boxes…
-
Thank you for releasing the codes of the CoCa3D and related dataset APIs in opencood.
**Issues**
When I trained the `dair_v2xvit.yaml` yaml file on DAIR-V2X-C_Example dataset with complemented annot…
-
Help!
When I input diffrent images with same point data from KITTI on MVXNet,The results has no change.
I have tried to find some reason:
Is there are some coding error on MVXnet config ?
in…
-
i cropped the training images so that the length and width are both 256.
my training code is as follows:
```
classes = {
'inner_distance': 1, # inner distance
'outer_distance': …
-
Hi Guys,
I guess there is an instability in the loss function due to a division by zero. can you please confirm that?
`def bbox_iou(box1, box2, x1y1x2y2=True, GIoU=False, DIoU=False, CIoU=False…
-
### Search before asking
- [X] I have searched the YOLOv8 [issues](https://github.com/ultralytics/ultralytics/issues) and [discussions](https://github.com/ultralytics/ultralytics/discussions) and fou…
-
## 🚀 Feature
A loss functions API in torchvision.
## Motivation
The request is simple, we have loss functions available in torchvision
E.g. `sigmoid_focal_loss` , `l1_loss`. But these are qu…
-
I trained my own model. Since generate_wts code you shared not use the config in DSVT repo, I have written my own "gen_wts.py"code.
Then, I deployed it from my own config. When I tried to inference…