A code gallery for person re-identification with mxnet-gluon, and I will reproduce many STOA algorithm.
Now, I only reproduce 5 papers, and train the model on marktet1501. I will add other datasets later.
I also will make further adjustments on these models. Before the end of August, all of these models and codes will be released.
1.Towards Good Practices on Building Effective CNN Baseline Model for Person Re-identification
2.In Defense of the Triplet Loss for Person Re-Identification(coming soon)
3.Beyond Part Models: Person Retrieval with Refined Part Pooling
4.Learning Discriminative Features with Multiple Granularity for Person Re-Identification
5.SphereReID: Deep Hypersphere Manifold Embedding for Person Re-Identification(coming soon)
Subscribe to Python Awesome
Get the latest posts delivered right to your inbox