triplet-reid

triplet-reid copied to clipboard

args.json file not available in Market-1501 TF Checkpoint Release

2

Please provide the 'args.json' file which was used during training in Market-1501 TF Checkpoint Release, so that 'embed.py' fiile could be used for generating embeddings.

zishansami

So this is a branch I had lying around for a while which adds a lot of things. I'm not merging it yet, as I still want to add these...

lucasb-eyer

Failed in training fine-grained categorization dataset CUB-200-2011

1

I use the same triplet loss (with BatchHard, Euclidean distance and Soft-margin) on the fine-grained categorization dataset [CUB-200-2011.](http://www.vision.caltech.edu/visipedia/CUB-200-2011.html) It aims to distinguish different species of birds (200 categorization, 5994 image...

JingyunLiang

Performance on CUHK03

10

Hello, authors. I was wondering, if you could provide some extra details about training on CUHK03. There is [third-party](https://github.com/Cysu/open-reid) re-implementation of your work. This implementation shows almost the same performance...

Edelbert

ValueError: Unable to configure handler 'stderr': bad argument type for built-in operation

4

I follow the README.md，but i can't run the train.py and the error is ValueError: Unable to configure handler 'stderr': bad argument type for built-in operation  can you tell me...

lihao056

Sensitivity to unclean datasets?

12

I've been doing some experiments with your batch hard triplet loss function and different architectures/datasets. On MARS I manage to reproduce the results from your paper (network seems to converge),...

dnaq

Questions about the re-implementation of lifted loss.

2

Sorry it's not the key point of your paper. After succesfully getting result of your batch hard loss function. I tried to check the result of lifted loss. When I...

ofishhaha

some question about triplet selection and training logs

7

Thanks your work firstly. I have some question want to ask you: (1) In your paper 3.3 , your set batch size to 72 containing 18 persons with 4 images....

zouxinh

Performance not as good on MARS [soln: combine by avg]

3

Hi, we have been doing some experiments to reproduce your results on the Market1501 and MARS dataset, and when using the exactly the same hyperparameters and training strategy in your...

YurongYou

Questions on using Inception_ResNet_v1 and test accuracy

4

Hi, I am new to deep learning, and thus may not understand your paper fully, hope is all right with you. I tried to implement the batch_hard using Inception_resnet_v1 and...

cptay