Lilin Chen

Is there complete training config setting proposed from paper?

1

您好, 想請教個問題 Paper中有提到作者您的 **Training Strategy** 想請問有沒有你們用來訓練imagenet的完整設定config檔可以套用參考呢? 感謝您~

[_Custom.cp38.pyd=>rotate_iou_matrix] calculate wrong value when boxes are exactly same.

1

## 🐛 Bug ## To Reproduce testing code: ``` from maskrcnn_benchmark import _Custom as _C from apex import amp rotate_iou_matrix = amp.float_function(_C.rotate_iou_matrix) boxes1 tensor([[672.4067, 290.7776, 791.0275, 38.9333, 34.1454]] boxes2 tensor([[672.4067,...

What's the behavior for segment "overlap" is true, but the dataset annotation exists some overlapping objects?

2

### Search before asking - [X] I have searched the YOLOv5 [issues](https://github.com/ultralytics/yolov5/issues) and [discussions](https://github.com/ultralytics/yolov5/discussions) and found no similar questions. ### Question For my understanding. Training the segment model we have...

**Describe the bug** I run the "demo/demo_attention_rpn_detector_inference.py.py" as example arguments, and it works. However, when I change the option "proto-net" to "baseline++" the error happened as follow. ``` File "C:\GitSource\SolVision3.0\AI_Vision_WPF\TaskProcess\bin\x64\Release\PYD\slmmm\packages\mmfewshot\demo\demo_metric_classifier_1shot_inference.py",...

Issue with SuperPoint and LightGlue Keypoint Matching

4

Hello, I've been utilizing the default SuperPoint for feature extraction combined with LightGlue for keypoint matching in this repository, and the results have been quite satisfactory in my use case....

請問支援prompt輸入嗎?

1

您好 首先感謝作者傑出的作品 我注意到你們conditional的條件是用類別來區分，不同於一般diffusion模型用text 且你們的模型輸入時也沒有prompt輸入 想請問你們的做法跟一般diffusion的差異 謝謝