Joint Discriminative and Generative Learning for Person Re-identification. CVPR'19 (Oral)

hi guys, thanks for your greate contributions. I am just curious while you do the supervised learning for MSMT17 datasets, did you reisze the images to the same size?(since the shape of the images vary in that dataset). If not, how did you deal with that?

Hello, in your ft_netAB id encoder defined in reIDmodel.py you detach the id_embedding:

f = self.model.partpool(x)
f = f.view(f.size(0),f.size(1)*self.
f = f.detach() # no gradient 

In the gen_update loss function in trainer.py you calculate the id embedding reconstruction loss (as specified in your paper) but you are using the detached embeddings:

self.loss_gen_recon_f_a = self.recon_criterion(f_a_recon, f_a) if hyperparameters['recon_f_w'] > 0 else 0
self.loss_gen_recon_f_b = self.recon_criterion(f_b_recon, f_b) if hyperparameters['recon_f_w'] > 0 else 0 

Because the embeddings are detached these losses are not constraining the model, did I miss something ?

Thank you for your code. I try to use MSMT and cuhk03 to train this model, but when I try to enter 'python train.py --config configs/latest.yaml' , command line output an error: ‘size mismatch for classifier.classifier.0.bias: copying a param with shape torch.Size([751]) from checkpoint, the shape in current model is torch.Size([2559]).' I have found out that I must change the param in latest-fp16.yaml and config.yaml(I think config.yaml may be useless in this issue). And I have found out that this error is from /models/best/net_last.pth. The input kinds isn't suitable for this net. Do you have any solution to this or should I change the file? @layumi

Have you tried cross-dataset testing - training on one dataset, say market1501 and testing on another, say cuhk03 or duke? Have you come across any model with code which has tried this? Will be really grateful to you, thanks in advance.

Hello, thank you very much for providing the implementation code of the DG-Net model. I encountered some problems during the implementation of the project. I will be honored if you can give me some suggestions.

I found the f=f.detch() in the ft_netAB. This causes the vector f to have no gradient. Then the loss_gen_recon_f_* has no no contribution to model parameter update. Is there something wrong with me or something wrong with the code?

I will be grateful if you can give me some suggestions. Thank you！

Hi Thanks for your great work. I use the pre-trained model for feature extraction using function “extract_feature” in the file “test2_label.py”. There is no problem with ‘.pt’ models in the “outputs” folder, but when I want to use the ‘net_last.pth’ model in folder “model” I receive the following error: KeyError Traceback (most recent call last) in () KeyError: 'a' Would you please guide me to debug this error? Is it about the “config.yaml” file exists in folder “outputs” and does not exist in folder “model”? Thank you very much and looking forward to hearing you Regards Hossein

https://github.com/NVlabs/DG-Net/blob/0abf564a853ea6ec3f38ab71a4a69f7f23b19d24/networks.py#L147

hey there, here the "outs0" is a feature matrix in shape of 8x1x64x32, and you used it to minus 1 and take the mean of them as loss. what is the principle behind this?

Hi @layumi , I was a bit confuse when read your code. I can't find appearance encoder and discriminator like your describe in your paper in your code. are they style encoder and MsImageDis in networks.py right ?

Hello, can you tell me if you have only implemented the change of appearance space, but not the transformation of structure space. For example, to generate a visual Angle image that does not exist in the structure space. Thank u

Hi, in DGNet_Trainer, you have gen_a, gen_b. I'm confused that in a single dataset training, do you need two generators a and b? What is the meaning of feeding image a and image b in your trainer? Thanks!

Defaults for this optimization level are: enabled : True opt_level : O1 cast_model_type : None patch_torch_functions : True keep_batchnorm_fp32 : None master_weights : None loss_scale : dynamic Processing user overrides (additional kwargs that are not None)... After processing overrides, optimization options are: enabled : True opt_level : O1 cast_model_type : None patch_torch_functions : True keep_batchnorm_fp32 : None master_weights : None loss_scale : dynamic Selected optimization level O1: Insert automatic casts around Pytorch functions and Tensor methods.

Defaults for this optimization level are: enabled : True opt_level : O1 cast_model_type : None patch_torch_functions : True keep_batchnorm_fp32 : None master_weights : None loss_scale : dynamic Processing user overrides (additional kwargs that are not None)... After processing overrides, optimization options are: enabled : True opt_level : O1 cast_model_type : None patch_torch_functions : True keep_batchnorm_fp32 : None master_weights : None loss_scale : dynamic 1500 265 1500 265 Every epoch need 188 iterations Note that dataloader may hang with too much nworkers. DLoss: 6.0000 Reg: 0.0000

Process finished with exit code 139 (interrupted by signal 11: SIGSEGV) I installed apex, and use the fp16 config, output this.

Hi, author. I have a question about your paper. You used online feeding and identity supervision to get the high quality of the generated image. Is it Primary feature learning and Fine-grained feature mining loss, respectively?

Hello,can you help me ?thank you! When I use Market data set ,report an error:

problem: /home/waq/anaconda3/envs/re_person/bin/python /home/waq/Downloads/ai/re_person/code/DG-Net/DG-Net-master/test_2label.py -------test----------- /home/waq/Downloads/ai/re_person/code/DG-Net/DG-Net-master/test_2label.py:182: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details. config = yaml.load(stream) Traceback (most recent call last): File "/home/waq/Downloads/ai/re_person/code/DG-Net/DG-Net-master/test_2label.py", line 204, in gallery_feature = extract_feature(model,dataloaders['gallery']) File "/home/waq/Downloads/ai/re_person/code/DG-Net/DG-Net-master/test_2label.py", line 131, in extract_feature x[0] = norm(x[0]) File "/home/waq/Downloads/ai/re_person/code/DG-Net/DG-Net-master/test_2label.py", line 109, in norm fnorm = torch.norm(f, p=2, dim=1, keepdim=True) File "/home/waq/anaconda3/envs/re_person/lib/python3.6/site-packages/torch/functional.py", line 1337, in norm return _VF.norm(input, p, _dim, keepdim=keepdim) # type: ignore IndexError: Dimension out of range (expected to be in range of [-1, 0], but got 1)

Process finished with exit code 1

Thank you！

Every epoch need 404 iterations Note that dataloader may hang with too much nworkers. Traceback (most recent call last): File "", line 1, in File "C:\Users\hcytech\anaconda3\envs\Python38_yolos\lib\multiprocessing\spawn.py", line 116, in spawn_main exitcode = _main(fd, parent_sentinel) File "C:\Users\hcytech\anaconda3\envs\Python38_yolos\lib\multiprocessing\spawn.py", line 125, in _main prepare(preparation_data) File "C:\Users\hcytech\anaconda3\envs\Python38_yolos\lib\multiprocessing\spawn.py", line 236, in prepare _fixup_main_from_path(data['init_main_from_path']) File "C:\Users\hcytech\anaconda3\envs\Python38_yolos\lib\multiprocessing\spawn.py", line 287, in _fixup_main_from_path main_content = runpy.run_path(main_path, File "C:\Users\hcytech\anaconda3\envs\Python38_yolos\lib\runpy.py", line 265, in run_path

Hello, When I try to train your model, I am getting the Runtime error:CUDA: out of memory error. Can you suggest me anything to get rid of this error?

Thank you.

Hi, author. I appreciate your work.

I am a new learner of reid. I want to do some vehicle reid with this project. The dataset i use is VeRi, whose structure is like this: ├── veri/ │ ├── image_train/ │ ├── image_test/
│ ├── image_query

我想知道如果训练车辆再识别，项目提供的teacher model和config配置文件等是否还可用？如果不可用，需要对项目做哪些修改。以后项目会不会更新车辆再识别的相关功能呢，请求您的指点。

thanks a lot.

In trainer.py, why only update the parameters of dis_a and gen_a and ignore the parameters of dis_b and gen_b? https://github.com/NVlabs/DG-Net/blob/a067be117b43c7c553275b6570b3a3bf8da465e0/trainer.py#L242-L248