alexattia / extendedtinyfaces Goto Github PK

I have difficulties in training your model?How to do that?Thanks for your great work and reply!

How do I change the code to make it work for video files. Please help me out here.

In tiny_faces_model.py file, you used a func which name is tf.image.extract_glimpse. It's to implement "context" in TINY FACES paper?

Is this any pretrained model

Hello alexattia:
The link of face recognition block was lost. Could you fix it? Thank you very much!

Thanks for sharing, but I still have some questions when running Video detection. it got a error when calling function tiny.evaluate:
TypeError: evaluate() got an unexpected keyword argument 'framee'
where tiny.evaluate(...) is defined in Tiny_Faces_in_Tensorflow/tiny_face_eval/ like follow style:

def evaluate(weight_file_path, data_dir, output_dir, prob_thresh=0.3, nms_thresh=0.1, lw=3, display=False):
No argument 'framee' indeed. So, you modified the original code ?
I will be even more grateful if you can help me.

If I want to detect other small objects
not only face
how should I do?
THANK YOU

Thanks for posting the paper and the code, stumbled upon this during my routine arxiv reading! Do you have any plans to release the pre-trained model?

Edit: Apparently I should have read through the code a bit more. It looks like you are using the pre-trained model from Tiny_Faces_in_Tensorflow