anhttran / 3dmm_cnn

Regressing Robust and Discriminative 3D Morphable Models with a very Deep Neural Network
Other
857 stars 254 forks source link

could not download the model: 3dmm_cnn_resnet_101.caffemodel,deploy_network.prototxt,mean.binaryproto #29

Closed emily1314 closed 6 years ago

emily1314 commented 6 years ago

Hi, thank for your sharing your paper and code, after reading the paper, I want to run the code, however, I could not download the caffemodel and prototxt. Can you tell me the link I can download the models? thank you so much.

hjy1312 commented 6 years ago

Hi, @emily1314 I have emailed you the link, please check it.

iacopomasi commented 6 years ago

The official link to download the CNN is following this page https://www.openu.ac.il/home/hassner/projects/CNN3DMM/

I tried now and it is all working.

bettertang commented 6 years ago

sorry , could not download the model now ,can you share the link again? thank you so much

anhttran commented 6 years ago

Can you try this? https://docs.google.com/forms/d/e/1FAIpQLSd6cwKh-CO_8Yr-VeDi27GPswyqI9Lvub6S2UYBRsLooCq9Vw/viewform Or: https://gist.github.com/iacopomasi/0646722f97fb087926a16b7e044ed6ed

Best, Anh

On Wed, Jul 4, 2018 at 6:53 AM, bettertang notifications@github.com wrote:

sorry , could not download the model now ,can you share the link again? thank you so much

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_anhttran_3dmm-5Fcnn_issues_29-23issuecomment-2D402485810&d=DwMCaQ&c=clK7kQUTWtAVEOVIgvi0NU5BOUHhpN0H8p7CSfnc_gI&r=uWuFHkuJNMJjKNj2bKpEDw&m=XgOUENIdMNrsClKNwEz-EuEZ0FzoPjD5moxyc5PkCx8&s=yfLcaLTRdOHIIZ3geAd9xPFInNAWsvt9CqThgZ7bSso&e=, or mute the thread https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_notifications_unsubscribe-2Dauth_AIpm-5FgNOq4QYPDP70h6eSSeQA4fgBvqkks5uDMjEgaJpZM4RSoiK&d=DwMCaQ&c=clK7kQUTWtAVEOVIgvi0NU5BOUHhpN0H8p7CSfnc_gI&r=uWuFHkuJNMJjKNj2bKpEDw&m=XgOUENIdMNrsClKNwEz-EuEZ0FzoPjD5moxyc5PkCx8&s=AGvTTvkukxWWpHMljh9nHMwILSnjzq17fliZklsKi_s&e= .

bettertang commented 6 years ago

thank @anhttran I can download the model now! But when i run python testBatchModel.py input.txt output/ in 3dmm_cnn/demoCode got the Error

: Prepare image data/1.jpg Traceback (most recent call last): File "testBatchModel.py", line 92, in dlib_img = io.imread(image_path) File "/usr/local/lib/python2.7/dist-packages/skimage/io/_io.py", line 62, in imread img = call_plugin('imread', fname, plugin=plugin, *plugin_args) File "/usr/local/lib/python2.7/dist-packages/skimage/io/manage_plugins.py", line 214, in call_plugin return func(args, **kwargs) File "/usr/local/lib/python2.7/dist-packages/skimage/io/_plugins/pil_plugin.py", line 35, in imread with open(fname, 'rb') as f: IOError: [Errno 2] No such file or directory: 'data/1.jpg\r'

how can I solve it ?

anhttran commented 6 years ago

Look like your "input.txt" uses Windows line ending "\r\n". You should re-download/re-create it in Unix to have Unix line ending ("\n"). Or you can drop the '\r' character at the end of image_path.

bettertang commented 6 years ago

extremely grateful @anhttran ,I run the demo and get the result

Zvyozdo4ka commented 9 months ago

The official link to download the CNN is following this page https://www.openu.ac.il/home/hassner/projects/CNN3DMM/

I tried now and it is all working.

No, it is not.

image image

anhttran commented 9 months ago

Dear all,

You could find my Caffe model in this link: https://drive.google.com/file/d/0BzkHndh8dgoATDZ5TnVRQk5qSkU

Best regards, Anh

Zvyozdo4ka commented 9 months ago

You could find my Caffe model in this link: https://drive.google.com/file/d/0BzkHndh8dgoATDZ5TnVRQk5qSkU

Thank You!

Is there any chance that you have your CNN model trained in PyTorch? I can't use this file: 3dmm_cnn_resnet_101.caffemodel, as I failed to install Caffe. I am trying to convert 3dmm_cnn_resnet_101.caffemodel into pth, but I haven't accomplished this yet.

anhttran commented 9 months ago

You can download the PyTorch model version, following the instructions in my another repo: https://github.com/anhttran/extreme_3d_faces#data-requirements.