kracwarlock/action-recognition-visual-attention
Action recognition using soft attention based deep recurrent neural networks
Jupyter Notebook
Issues
- 3
Results reproducibility
#13 opened by jacopocavazza - 2
How to get the dataset?
#15 opened by cumtwb - 3
Can anyone share the code mentioned in the closed issue #6 which is used to combine features to h5 format?
#20 opened by YinRui1991 - 0
I'm new to this task. Could I know that how many frame are feed in testing time?
#21 opened by guotong1988 - 23
- 1
Python Scripts for CNN Encoding Generation
#19 opened by 345ishaan - 0
Possible bug on data_handler : Reset()
#18 opened by noky - 0
- 5
Your time for one 128-batch?
#16 opened by thanhnguyentang - 4
Location softmax
#12 opened by pmorerio - 0
Multi-layer LSTM
#14 opened by pmorerio - 21
Initialization of LSTM layers
#3 opened by xlliu7 - 1
ucf11 evaluation problem
#11 opened by ae86208 - 2
- 1
Possible redundant code in actrec_mAP.py
#9 opened by nixingyang - 1
Enquiries regarding Data Preprocessing
#8 opened by nixingyang - 1
- 10
Input data
#2 opened by frajem - 1
Best model selection
#4 opened by frajem - 0
1
#5 opened by YantianZha - 3
How to get the h5 features file
#1 opened by Yidann