zlmzju / part_reid Goto Github PK
View Code? Open in Web Editor NEWCode for ICCV2017 paper: Deeply-Learned Part-Aligned Representations for Person Re-Identification
Code for ICCV2017 paper: Deeply-Learned Part-Aligned Representations for Person Re-Identification
From the readme - it seems that the train.sh script is missing in the train folder. I am looking for if it is possible to get the training script for replicating and improving by tuning hyperparameters.
Hi!
Do you have or can you recommendation any pytorch or tensorflow implementation of your paper?
Thank.
I am try to reproduce the results on market1501. But the triplet loss is very small, nearly to zero. So after iter=35000+, the network did not update(loss=0). I can't reproduce the results(R@1:81%,mAP:63%), but get lower results (R@1:76%,mAP:55%).
Have you met this situation? What's the possible reasons ?
make: *** [.build_release/src/caffe/layers/online_triplet_loss_layer.o] Error 1
src/caffe/layers/online_triplet_loss_layer.cpp: In member function \u2018virtual void caffe::OnlineTripletLossLayer::Forward_cpu(const std::vector<caffe::Blob>&, const std::vector<caffe::Blob>&)\u2019:
src/caffe/layers/online_triplet_loss_layer.cpp:100:27: error: \u2018>>\u2019 should be \u2018> >\u2019 within a nested template argument list
vector<pair<Dtype, Dtype>> min_max_distances;
^
src/caffe/layers/online_triplet_loss_layer.cpp: In instantiation of \u2018void caffe::OnlineTripletLossLayer::Forward_cpu(const std::vector<caffe::Blob>&, const std::vector<caffe::Blob>&) [with Dtype = float]\u2019:
src/caffe/layers/online_triplet_loss_layer.cpp:287:1: required from here
src/caffe/layers/online_triplet_loss_layer.cpp:70:13: warning: unused variable \u2018dim\u2019 [-Wunused-variable]
const int dim = bottom[0]->count() / num;
^
src/caffe/layers/online_triplet_loss_layer.cpp: In instantiation of \u2018void caffe::OnlineTripletLossLayer::Forward_cpu(const std::vector<caffe::Blob>&, const std::vector<caffe::Blob>&) [with Dtype = double]\u2019:
src/caffe/layers/online_triplet_loss_layer.cpp:287:1: required from here
src/caffe/layers/online_triplet_loss_layer.cpp:70:13: warning: unused variable \u2018dim\u2019 [-Wunused-variable]
Makefile:581: recipe for target '.build_release/src/caffe/layers/online_triplet_loss_layer.o' failed
make: *** [.build_release/src/caffe/layers/online_triplet_loss_layer.o] Error 1
Firstly, thanks your great work!
but i cannot use your python layer to train my model. it will stop when cv2 reading one image.
could you give me some help about this?
thanks
In the extract_features method in the Extractor() class, there is a line:
normed_features=self.net.blobs['normed_feature'].data.copy()
I checked the _caffe.cpp file in ./caffe/python/caffe/_caffe.cpp and I then checked all the properties for net.blobs and 'normed_feature' was not one of them.
Any ideas on how to resolve this?
Thanks
Hello, I want to down load the dataset. But there is a warning that "this page is asking you to confirm that you want to leave - data you have entered may not be saved". How can I down load the dataset?
how about your gallery and probe ?
and training data
Hi @zlmzju,
What is the license?
Thanks
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.