Awesome
code for Searching Actions on the Hyperbole. <br>
The current code includes all the essentials to re-implement our paper, <br> The demo code is a raw version, please leave comments if there is any difficuty to use it. <br>
Requirements
pytorch == 1.4.0 <br> geoopt == 0.2.0 <br>
If you fail to use this code, please email me (uestc.longteng@gmail.com) <br>
Data Hierarchy and Seen/Unseen Split.
activity_net_depth.csv <br> kinetics_depth.csv <br> moments_depth.csv <br>
Data files
I have uploaded the extracted video features (~8GB) to Microsoft Onedrive, please contact me for data sharing.
Pretrained 300-D action embedding:
cone_activity_net300.pth <br> cone_kinetics_300.pth <br> cone_moments_300.pth <br>
Reference
Searching for Actions on the Hyperbole
Teng Long, Pascal Mettes, Heng Tao Shen, Cees G. M. Snoek
@InProceedings{Long_2020_CVPR, <br> author = {Long, Teng and Mettes, Pascal and Shen, Heng Tao and Snoek, Cees G. M.}, <br> title = {Searching for Actions on the Hyperbole}, <br> booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}, <br> month = {June}, <br> year = {2020} <br> }