search

liudaizong / CSMGAN

Code for ACM MM2020 paper: Jointly Cross- and Self-Modal Graph Attention Network for Query-Based Moment Localization
33 stars 5 forks source link

about kernel size #2

Open xinfangliu opened 3 years ago

xinfangliu commented 3 years ago

"we adopt convolution kernel size of [16, 32, 64, 96, 128, 160, 192] for Activity Caption, and [8, 16, 32, 64] for TACoS." actually, the kernel size is 1 in your code. This may not be called a multi-scale window. In addition, could you please provide your results using the correct IoU calculation?

SCZwangxiao commented 3 years ago

Did you succeed in reproducing the results? I failed in both Activitynet-caption and TACoS dataset, using default settings in the code.