Comments (38)
过几天我加一个MTCNN的检测接口你再试试~
from ncnn_example.
@1976277169 好的,多谢~ 。
from ncnn_example.
我也想尝试一下,把左博的那个mtcnn加进来试试
from ncnn_example.
来来来,可以把左博所有的模型转一波,再出一个项目~
from ncnn_example.
要跟着你学学,哈哈,我知识层面还太窄了
from ncnn_example.
要跟着你学学,哈哈,我知识层面还太窄了
跟左博学习吧,认认真真的啃ZQCNN~
from ncnn_example.
可以试一下重新上传的大模型,可能效果会好一点~
from ncnn_example.
可以试一下重新上传的大模型,可能效果会好一点~
能否把输入图像尺寸变大一点?比如,224x224, 或416x416, 效果应该会好一点?
from ncnn_example.
可以试一下重新上传的大模型,可能效果会好一点~
能否把输入图像尺寸变大一点?比如,224x224, 或416x416, 效果应该会好一点?
网络输入是固定的,如果修改网络输入大小,需要修改模型结构,最后很可能不能实时~
from ncnn_example.
96x96的效果, 大佬,我这效果也太差了, 你测试的有这么差吗?
from ncnn_example.
可以将你的测试代码上传一下吗?我这边测试没这么渣,这个模型应该是比之前上传的模型要好的~
from ncnn_example.
代码:
https://github.com/taylorguo/MTCNN_Landmark106
from ncnn_example.
ncnn::Mat in = ncnn::Mat::from_pixels_resize((const unsigned char*)indata, ncnn::Mat::PIXEL_BGR, width, height, 48, 48);
LOGI(" ******** JNI 图像格式转换完成, Mat已读入, 大小已经更改为(96x96).");
这个是你最终的测试代码吗?
你需要将这里的48,48改成96,96~
from ncnn_example.
from ncnn_example.
java代码看着有点晕,传入关键点检测的image需要是对应人脸区域,也就是说对人脸检测结果进行关键点定位,你这里关键点检测传入的是人脸检测结果吗?
from ncnn_example.
是的,我用https://github.com/moli232777144/mtcnn_ncnn
把人脸方框提取出来, 然后把方框裁剪出来放到网络里面推断
from ncnn_example.
你将人脸检测框也画上去看看~
from ncnn_example.
from ncnn_example.
48x48的模型测试过了吗?
from ncnn_example.
测过了,上面发的第一个测试图是48X48的,下面这个是96x96的
难道是因为图像通道数有问题?
Android里面传进来的是4通道的
from ncnn_example.
你的测试结果发一下看看呗?
from ncnn_example.
给我一张你的测试图片,我晚上测试一下给你结果~
from ncnn_example.
from ncnn_example.
应该是from_pixels_resize()第二个参数出问题了,改成PIXEL_RGBA2BGR,稍微好一点,但还是有问题
你帮我看看怎么改?
我还是要仔细看看ncnn源代码
from ncnn_example.
那就是你代码有问题咯,你仔细对着改吧,我这边模型肯定是没问题的,都测试过的~
from ncnn_example.
这个参数应该是PIXEL_RGB2BGR
from ncnn_example.
你可以参考下我给的测试代码~
from ncnn_example.
你的测试效果,你还是帮我测一下,发我一下,我看看我能优化到什么程度
from ncnn_example.
不好意思,不会java,这个我心有余而力不足我最多在linux或者windows上测试一下
from ncnn_example.
是的,你在其他系统上测试的结果
from ncnn_example.
图像的通道已经改成3通道了,还是有问题,检测出来的landmark分布和上面这张图一样
from ncnn_example.
我也是有点晕了,难道非要导个opencv的包进去??
from ncnn_example.
from ncnn_example.
模型出来的点坐标没问题,Java画点有问题.
from ncnn_example.
推断的时候,能否改成接受灰度图?
from ncnn_example.
不接受,你重新提个issue吧,这个已经关闭了,我新加了retinaface做人脸检测,可以去测试一下~
from ncnn_example.
retinaface你是转的哪个项目的
from ncnn_example.
有问题,你可以重新开一个issue,找半天才找到retinaface转的是insightface的吧,忘记了,很久以前的事了
from ncnn_example.
Related Issues (20)
- /usr/bin/ld: cannot open output file face: Is a directory HOT 7
- MacOS Catalina 10.15.5 error: unsupported option '-fopenmp' HOT 4
- Are there any good Gender&Age pretained models Available on ncnn? HOT 2
- mobilefacenet的模型是从哪个原始模型转过来的 HOT 7
- How to get landmark from "RetinafaceFactory" HOT 3
- Can not build cmake, error can not find -lncnn HOT 2
- SigArt error HOT 3
- mask detection error HOT 9
- 编译错误 HOT 2
- 大佬,能问下这个分类用的模型mobilenet.param是什么转过来的嘛? HOT 1
- Detect Mask? HOT 2
- not support the latest NCNN? HOT 4
- How to calculate the similarity(score) of arcface in the insightface github? HOT 1
- Destination points in aligner HOT 1
- Design patterns HOT 1
- Do you use quantize model when you convert to ncnn format ? HOT 3
- 建议说明模型出处 HOT 2
- 人脸对齐关键点有问题? HOT 14
- How to install for linux? HOT 1
- windows:Microsoft.CppBuild.targets(436,5): error MSB8013: 此项目不包含配置和平台组合 Debug|x64 HOT 22
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_example.