SCDM
SCDM copied to clipboard
Code for the paper: Semantic Conditioned Dynamic Modulation for Temporal Sentence Grounding in Videos
Hello, The links [charades_i3d_rgb.hdf5](https://cloud.tsinghua.edu.cn/f/fd155bd64baf4359ada9/?dl=1), [activitynet_c3d_fc6_stride_1s.hdf5](https://cloud.tsinghua.edu.cn/d/203f17ed08ae41f69a0e/), [tacos_c3d_fc6_nonoverlap.hdf5](https://cloud.tsinghua.edu.cn/f/876b5550bd884bf1bb77/?dl=1) are no longer available. Would you mind making them public again? Thanks!
Hello, very meaningful work. How many GPUs are needed for the training? How long is it altogether?
Hi, I am trying to use your code to generate training and testing data of Tacos, but the "timestamps" and "duration" in diractionary of three JSON files confused me. It...