Comments (16)
R50模型转换可以看下有没有不支持的层,如果有的话可能不能直接转。需要自己写模型转换工具。
from retinaface.
@clancylian 我转出来,在caffe上跑,结果不一样,面部框变大了,而且5个关键点也不准确。你转出来的结果和官方一样嘛?
from retinaface.
@xiakj @clancylian
我这边实验也是,代码没改,mnet0.25跑出的结果没有问题;但是resnet50转过来的caffemodel,检测不到人脸。这是什么情况呢??
from retinaface.
resnet50的转换方法参考:https://github.com/szad670401/advanced-mxnet2caffe
from retinaface.
@OPPOA113 现在检测结果正确了,一定要保证测试图像比例不要发生变化,可以通过调整*_deploy.prototxt中的data层的dims参数解决,也可以在源码中设置。
from retinaface.
@xiakj 谢谢您。
请问一下,
1、你用的是哪个工程转的caffemodel?
2、这个图像比例设置的是多少,参考的是哪里的?
from retinaface.
- 就是上面你提到的那个advanced-mxnet2caffe。也有已经转换号的参考https://github.com/Charrin/RetinaFace-Cpp
- 我输入图像是1280720的,因为我的gpu性能比较低,所以我按照比例将dims设置为360640了
from retinaface.
@xiakj @clancylian
采用这个工程下的,通样的代码,图像都是resize到1280x720的输入,mnet0.25检测的结果正常,RestNet50检测的结果不正常,这可能是什么地方出问题了?
from retinaface.
@OPPOA113 不太了解,感觉RestNet50模型本身的问题吧,你再好好读读RestNet50的prototxt
from retinaface.
@xiakj
你用的R50的caffemodel和prototxt能不能发过来给我。我试试。[email protected]
from retinaface.
@OPPOA113 请查收
from retinaface.
@xiakj 谢谢,已收到。
但是检测的结果还是不对。 mnet0.25 模型检测是没有问题的,不知道resnet50检测结果为何不正确.......
from retinaface.
@xiakj @OPPOA113 我使用https://github.com/Charrin/RetinaFace-Cpp下方法转换后的R50,也是无法检测出来。请问二位解决这个问题了吗?能否发送我一份模型和prototxt,谢谢[email protected]
from retinaface.
@nuanxinqing 您可以试一下github.com/szad670401/advanced-mxnet2caffe,我是用这个转换出来的。但是caffemodel转换过程中也遇到了一些问题,您留意一下,也可以再沟通。
from retinaface.
这里 https://github.com/cholihao/Retinaface-caffe 找到了一个google drive链接, 里面有caffe的模型: https://drive.google.com/drive/folders/1VoABSiHXiVlRCEryKtf3UG_BW9236UUp
from retinaface.
@jiapinai 已经好了,谢谢您!
from retinaface.
Related Issues (20)
- mxnet model HOT 2
- Build Error HOT 5
- resizeconvertion.cu HOT 2
- mxnet的速度是用官方insightface代码测试出来的吗?还是自己做过优化? HOT 1
- 请问你这个代码比官方的快吗?
- building on mobile
- none
- 用pytorch转换的onix模型,可以用你的代码调用吗?
- tensorrt version and opencv version? HOT 2
- what's your tensorrt version?
- 怎样才能把caffe模型修改成可以tensorRT的格式? HOT 1
- how can i build on windows HOT 1
- retinaface res50 have alittle question
- 使用caffe时能否设置多GPU同时工作?
- Batchsize support for image processing with Cuda
- tensort模型检测结果与caffe模型相差较大 HOT 1
- cudnn_conv_layer.cpp:56] Check failed: error == cudaSuccess (2 vs. 0) out of memory
- Why the test time when using USE_CAFFE and USE_TENSORRT is almost same?为什么测试时间差不多
- mnet-deconv-0517.table.int8 only 7.6KB
- Docker for this project
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 retinaface.