Comments (6)
Good question. I recommend training all 25 templates on every scale. My experience is that training templates only on corresponding scales performs slightly worse.
from tiny.
Thanks peiyun. I just want to reproduce the result of your paper base on the py caffe. But now it seems that the result has something wrong. I think I have achieved most of ideas of your paper except more batch size and ohem. But the mAP in WiderFace validation set is only 6%. Too many false positives and bad performance for easy face. Could you give me some suggestions about that? Hope to see your training code as soon as possible. Thank you!
from tiny.
Hello peiyun, I found that if I reuse the parameter in resnet between ohem layer and hard mining layer subsequently , the AP value drop down about 10 percent t. But all the ohem paper suggest to reuse parameter .
Another problem is should I put the use_global_stats to be true or false of bn layer in training? It seems that how to implement the ohem in bn layer in resnet is very different with vgg16.
from tiny.
We did not implement the online hard negative mining in the same way as the OHEM paper did. Please refer to "Online hard mining and balanced sampling" in the appendix of our paper for our implementation.
from tiny.
Hello peiyun, thanks for your clarification. I have another question is how to " adding context" in resnet network? Firstly , should we use concat layer or Eltwise layer to adding context? Secondly , when I try to adding conv2, conv3 and conv5 , I found that the value of conv5's feature map in resnet is much larger than conv2 and conv3. If I adding these layers together, it seems that the featuremaps of conv2 and conv3 can not take any effect on the final result. Should I scale the value to match conv5? Thanks!
from tiny.
Here is a visualization of our network: http://ethereon.github.io/netscope/#/gist/8a0d5ef37da9dc4cd611d178404b3641. Hopefully it will address your questions.
from tiny.
Related Issues (20)
- How to train with my database? HOT 4
- How to pre-process the training data HOT 1
- hr_res101('train') error: vl_argparse error HOT 1
- How to speed up the process to use it with webcamera?
- Which layer can we use for face recognition params?
- Where is ellipse linear regression?
- how to convert the text file to mat file like wider face ?
- Question on train detector on my own dataset HOT 1
- Training on own data HOT 1
- Initialise Resnet model HOT 1
- Foveal descriptor HOT 1
- too much training time
- I have the same question too. It confuse me a few days. I read part of the code, it seems that the output 'score_cls' is a 25 depth matrix or tensor which corresponding to each 'templates'. I just don't understand the why?? If someone know it, I'm very thanksful to you. HOT 2
- hr_res101.mat broken link HOT 2
- Can not download model HOT 1
- new issue
- How can we use our model to detect faces with camera in real time? HOT 1
- About the function 'nms_mex' error HOT 1
- nvcc fatal: Unsupported gpu architecture 'compute_86'
- Links to pretrained weights broken? HOT 1
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 tiny.