ncnn-android-yolov8

ncnn-android-yolov8 copied to clipboard

你好！大牛！ 能否提供下如何导出 yolov8-obb 到ncnn模型的具体步骤，谢谢！！ 期待您的回复！！

zgsxwsdxg

在Android，已经运行了，如果用在harmony，该怎样做？谢谢

lizhangsongzi

generate_proposals seg faults on line 155 with a single class model

2

When running with a single class model, generate_proposals will seg fault on line 155. It's doing so when i=1958. The dimensions of pred are w=8,400 and h = 5. Total...

ChthonicOne

Android studio 识别不出来这是一个项目，点击run时只运行一，没有将app安装到模拟器上？

4


Liusdaooo

Input Image insted of liveCamera

7

Could you help me incorporate an image input into the model, resulting in an output that includes bounding boxes? i tried like this but not wroked MainActivity=> public void PassImage()...

ISTIAKPAPPU

Anythimg work fine until i set the view orientation to landscape ,it will make preview rotate 90 degree. I have tried setting orientation = (int)e.data.i32[0]; ->orientation = (int)e.data.i32[0]- 90 ;...

chenchen-boop

更换为自己的模型之后闪退

14

Logcat报错部分日志为： 2023-11-26 17:54:03.350 7736-7749 ACameraDevice com.tencent.yolov8ncnn D Device error received, code 5, frame number 0, request ID 0, subseq ID 0 2023-11-26 17:54:03.353 7736-7750 ACameraDevice com.tencent.yolov8ncnn D Device error received,...

LHT130

按照教程修改了C2f和Detect模块，也修改了yolov8-obb.cpp里的num_class和标签数组 

XTYCoding

训练自己数据集后,生成onnx文件,转ncnn部署到手机上没有检测出目标

2

用官方提供的没有问题,自己训练数据集没有更改网络信息也没有检测框。 yolov8ncnn.cpp 文件中的target_sizes 设置成640*640可以检测，但是设置成320*320就是检测不到目标物体。官方的模型导出来的ncnn模型就没这个问题，随便怎么设置，都能检测到

superspeedone

Hi, I have been able to successfully build run the base yolov8 models. Even custom models work well. However, I am not too sure how to use the segmentation model...

vjsyong