-
请问作者我的非极大值抑制过程中总是有400个维度并且其中大多数都是重复的,导致打印标签的时候总是有两百个标签,但是标签都一样
-
## error log | 日志或报错信息 | ログ
base) ➜ bin git:(master) ./pnnx end2end.pt
pnnxparam = end2end.pnnx.param
pnnxbin = end2end.pnnx.bin
pnnxpy = end2end_pnnx.py
pnnxonnx = end2end.pnnx.onnx
ncnnparam…
-
I use Google colab to run the demo.py,but it shows that No module named 'nms.gpu_nms'.How can I deal with this problem?
![image](https://user-images.githubusercontent.com/39090271/46515236-2b7b8e00-c…
-
The latest version replaces nms with soft-nms
-
Allowing training on MSP device besides CUDA and CPU.
**Motivation**
Since the introduction of PyTorch Nightly, it is now possible to use tell PyTorch to use the embedded GPU of the new M1 chips o…
-
### Your Environment
- **Version:** V1.9
- **Affected Component:** Orchestrator, Access Gateway, or NMS
### Describe the Issue
We are facing below error while testing V1.9:
- Orchestrator Ve…
-
Hello,
Is there a way to activate or implement the NMS clustering on overlapping detections belonging to different classes. I want to eliminate the problem of the same object being detected as both…
-
I test the code,it seems that nms does not work.The effect is shown in the figure below:
and can not detect object in far distance.Is anybody know how to fix it? Thank you!
![image](https://user-ima…
-
@1996scarlet Hi, thanks for your sharing. would you mind sharing the details of the difference between the optimized NMS and original NMS? Thanks a lot.
-
Hi,
In the Stardist3D paper, it is explained that 3D NMS is performed efficiently by computing a cascade of tighter and more computationally intensive bounds on the IoU between two candidates. The ca…