farquasar1 / convlstm Goto Github PK

Hi, thanks for sharing the code.
where is the data?

I can't find the package ugs_utils, and when I run the code, I get the output like this

ImportError: No module named data_preprocessing

which is line 12 in lstm_train_fcn.py.

Hi, thanks for the code.
I'm now trying to run the lstm_train_fcn.py script and I have some questions:

1. I have 2 video sequence named as train.mp4 and train_label.mp4, but I don't know which folder should I put these two videos?
2. I got an error FileNotFoundError: [Errno 2] No such file or directory: 'splits.txt'. How should I prepare this 'splits.txt'?
   I'd be really appreciated if you can provide some help.

Hi there,
how do you install "hal" for data augmentation? it is imported like this,

from hal.augmentation.geometric_transforms import random_rigid_transform, apply_geometric_transform

please I want to ask the file "splits.txt' .what is this ?

How to use this convLSTM project for object detection inside a video dataset?

Hi @farquasar1 , thanks for the code sharing.

I tried to implement this network to do some 3d image segmentation, my input shape is (80,40,40,3),
which 80 is frame size, 40 and 40 are rows and col, 3 is channels. After I try to run the training, I encounter this problem with the first time distributed(maxpooling) layer.

ValueError: Negative dimension size caused by subtracting 2 from 1 for 'time_distributed_2/MaxPool' (op: 'MaxPool') with input shapes: [?,20,1,24].

Thanks for your time! :D