danielelic / deep-segmentation Goto Github PK

On data.py, lines 58-59

    images = np.array(images, dtype=np.uint8)
    images16 = np.array(images, dtype=np.uint16)

Both are using the 8-bit images, so the training will not use the 16 bit resolution.

Is it so?

Hi Daniele, thanks again for your code, it's been very useful.

I'm wondering how I can articulate this code with video files in more traditional formats such as .mov, .avi or .mp4. Is it possible to convert these videos to .oni? What's the origin of an .oni video? Or do you have any other idea that could help me?

Thanks!

https://structure.io/structure-sensor/getstarted

Hi Daniele,

I'm trying to train the model according to the "Run" section but when I run the command python train_unet3_conv.py I get this:

(...)

Fitting model (bit = 8) ...

Train on 0 samples, validate on 0 samples
Epoch 1/200
Traceback (most recent call last):
File "train_unet3_conv.py", line 232, in
train_and_predict(8)
File "train_unet3_conv.py", line 186, in train_and_predict
callbacks=[csv_logger, model_checkpoint])
File "C:\Program Files\Anaconda3\lib\site-packages\keras\engine\training.py", line 1039, in fit
validation_steps=validation_steps)
File "C:\Program Files\Anaconda3\lib\site-packages\keras\engine\training_arrays.py", line 217, in fit_loop
callbacks.on_epoch_end(epoch, epoch_logs)
File "C:\Program Files\Anaconda3\lib\site-packages\keras\callbacks.py", line 79, in on_epoch_end
callback.on_epoch_end(epoch, logs)
File "C:\Program Files\Anaconda3\lib\site-packages\keras\callbacks.py", line 338, in on_epoch_end
self.progbar.update(self.seen, self.log_values)
AttributeError: 'ProgbarLogger' object has no attribute 'log_values'

Apparently there's no samples to train or to validate, but the npy folder is (i guess) well created.

How can I solve this? is there something I'm missing?

Thanks a lot

How do I link the libOpenNI2.so and the OpenNI2 directory?

Hi:
thank you very much for opening the source code . can you share your training model and paper.