-
Hi. Thank you for generously sharing your work. When I trained the model on MSRVTT with 1 V100, I found the GPU-Util cannot reach 100%(about 60%). Do you have some tips? Thank you.
-
Hi,
Nice work! I understand HERO is best designed for multi-channel tasks, but I am curious about how HERO performs on How2QA and TVQA when not using the subtitle. Do you have these? It'd be helpful …
-
Hey @tgc1997
Thanks for providing the implementations of such an awesome work!!!
I wanted to know how does one go about using the pre-trained models for inferencing on raw custom videos?
-
In your link, the dataset MSR-VTT is in mp4 however in your code the formation of videos should be in hdf5. Does it mean I need to change the mp4 video into hdf5 firstly?
-
Hi,
When I training more epochs in mar-cry, it seems much better. And why do you 5 epochs.Thanks.
-
it is nice to read your paper and I find this work is very interesting to me
I am wondering when will you release your data and code?
-
when I test your model on youcook2 data, the performance is not very good.
I am not sure what is the main reason behind it. have you conducted similar experiments before?
youcook2 is longer than…
-
Hi,
When I run Clip4Clip code to exact video frames, this code in rawvideo_utils.py:cap.set(cv2.CAP_PROP_POS_FRAMES, sec_base + ind),print some warning message like:
[h264 @ 0x56543b236ac0] illegal …
-
Hello, yiskw713:
I am rebuilding your repo and during the rebuilding, I was confused about some concepts in ```config.yaml```.
First, is ```dataset_dir``` means the dir of features extracted by …
-
Hey,
Thanks for releasing the codes!
I am trying to train MSRVTT retrieval, using this line:
`python src/tasks/run_video_retrieval.py --config src/configs/msrvtt_ret_base_resnet50.json …