Walid Ahmed

Hello, How can the FCN responsible for predicting masks deal with the different dimensions of the ROI? it seems the output of masking ROI is 14*14*number of classes How can...

Getting low accuracy

1

I always end up with accuracy about 0.5 and loss of about 1. I assume these value are so low What would be the default values to expect from SSD300?

Using SSD500?

1

Hi, is there is a way I can used SSD500 rather than SSD300? I could nod find the architecture for SSD500 Thanks

In the original paper the neg_pos_ratio was equal to 3, the current implementation uses 2 Why was this change assumed to be better? did anyone try to play around with...

Minimum size for an object to detect

3

Hi, What would be the minimum size for an object to detect on the 300px by 300 px ? Thanks

TypeError: Expected int32, got list containing Tensors of type '_Message' instead.

2

I am having the following error model = SSD300(input_shape, num_classes=NUM_CLASSES) File "/Users/walidahmed/Desktop/Code/scripts/ssd.py", line 258, in SSD300 mode='concat', concat_axis=1, name='mbox_loc') File "/usr/local/lib/python2.7/site-packages/keras/engine/topology.py", line 1669, in merge name=name) File "/usr/local/lib/python2.7/site-packages/keras/engine/topology.py", line 1290,...

question on training with other dataset

2

@lvaleriu Hi your code worked fine when I applied it to images, video and cam. I am trying to classify vehicles and pedestrians, I checked the file gt_pascal.pkl and read...

Thanks a lot for great code, I still have some questions Does it use SVM? What is the “peak value” returned and what is its range? where actually it look...

Can not load TF model after converting

2

Hi All I converted the model successfully and without any error message, I also created the pbtxt file. However when I try to load the model using OpenCV I am...

Should not the input_shape = (16,112,112,16) instead of input_shape = (112,112,16,3) ?