Comments (6)
I just submitted a pull request for yolov8 example #5506 I hope it can be helpful to you.
Hi, I tested the yolov8.cpp example in your PR, but the results are unreasonable. Specifically, I got yolov8n.bin and yolov8n.param following your guide in yolov8.cpp, and your yolov8n example returns many detections with same confidence of 0.5. The result:
我知道这个问题,这个原因不是因为我这边的原因,而是您导出的yolov8.ncnn.* 模型是仅固定支持[1,3,640,640]的模型,并不支持动态shape输入造成的。
这个错误的模型会导致输入进去的数据为了保持图像原比例缩放,并不符合模型要求的[1,3,640,640],实际上有可能是[1,3,640,540],所以模型内部会错位读取图像,对结果造成严重偏差。
只是要尝试效果的话,您可以直接把输入的图片用画图工具或者cv::resize() 修改成640x640的尺寸。
pnnx本身是支持动态shape输入的,您可以去参考 https://github.com/pnnx/pnnx?tab=readme-ov-file#------detailed-options
导出正确的支持动态shape输入的模型,期待您的好消息~🤣
Well, I find out that the problem originates from a bug in your yolov8.cpp
the following lines:
int wpad = (w + MAX_STRIDE - 1) / MAX_STRIDE * MAX_STRIDE - w;
int hpad = (h + MAX_STRIDE - 1) / MAX_STRIDE * MAX_STRIDE - h;
should be modified as:
int wpad = (target_size + MAX_STRIDE - 1) / MAX_STRIDE * MAX_STRIDE - w;
int hpad = (target_size + MAX_STRIDE - 1) / MAX_STRIDE * MAX_STRIDE - h;
from ncnn.
wait for
from ncnn.
I just submitted a pull request for yolov8 example #5506
I hope it can be helpful to you.
from ncnn.
I just submitted a pull request for yolov8 example #5506 I hope it can be helpful to you.
Hi, I tested the yolov8.cpp example in your PR, but the results are unreasonable. Specifically, I got yolov8n.bin and yolov8n.param following your guide in yolov8.cpp, and your yolov8n example returns many detections with same confidence of 0.5. The result:
from ncnn.
I just submitted a pull request for yolov8 example #5506 I hope it can be helpful to you.
Hi, I tested the yolov8.cpp example in your PR, but the results are unreasonable. Specifically, I got yolov8n.bin and yolov8n.param following your guide in yolov8.cpp, and your yolov8n example returns many detections with same confidence of 0.5. The result:
我知道这个问题,这个原因不是因为我这边的原因,而是您导出的yolov8.ncnn.* 模型是仅固定支持[1,3,640,640]的模型,并不支持动态shape输入造成的。
这个错误的模型会导致输入进去的数据为了保持图像原比例缩放,并不符合模型要求的[1,3,640,640],实际上有可能是[1,3,640,540],所以模型内部会错位读取图像,对结果造成严重偏差。
只是要尝试效果的话,您可以直接把输入的图片用画图工具或者cv::resize() 修改成640x640的尺寸。
pnnx本身是支持动态shape输入的,您可以去参考 https://github.com/pnnx/pnnx?tab=readme-ov-file#------detailed-options
导出正确的支持动态shape输入的模型,期待您的好消息~🤣
from ncnn.
close for #5506
from ncnn.
Related Issues (20)
- 在mt8786 (Cortex-A55*6 Cortex-A75*2) 上推理int8模型速度提升不明显 HOT 1
- 对特定网络层进行推理配置时,执行layer->create_pipeline方法时crash
- 请求c_api.h里添加set_vulkan_device方法用于指定GPU设备
- 小米12 Pro无法使用64位库 HOT 3
- quantized torch/onnx model conversion into ncnn model
- crash while load param
- 请问pnnx转换是否支持nn.GRUCell呢?
- windows 10 使用 vulkan 利用 gpu 加速推理的时候,每次退出都会崩溃。0x00007FFB6D8A1515 (nvoglv64.dll) HOT 2
- ncnn::Mat::from_pixels导致内存不稳定
- 在Ubuntu 24.04 TLS 中使用 gpu 推理时返回 -nan
- “layer Resize not exists or registered” Should I register an ops?How can I add a new ops? HOT 1
- 使用最新版的pnnx,pytorch实现的一个模型里面用了一个全连接层nn.linear(),在ncnn会转换为F.linear,但是报错ncnn没有该层。 HOT 1
- Add signature or change default extension of weights files HOT 7
- GiHub actions for Android build is failing HOT 2
- 新加的examples/yolov8.cpp未对c++98兼容
- 现在ncnn支持pytorch的unfold命令吗,我运行报错 HOT 2
- ncnn benchmark跑模型为什么cpu占用都是满的
- clang-18 ubuntu24.04 x64 buidl fail
- HarmonyOS cpu.cpp get_thread_siblings 返回错误 HOT 1
- Yolo v5 not showing Object detection Prediction Results with GPU on RISC-V. HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from ncnn.