Khoa Vo
Khoa Vo
@sharmalakshay93 @ay27 If you're using annotation files from COCO 2017. You can add a new key namely 'type' with any value except 'instances'to the json file to resolve this problem,...
I have also met this problem and have to replace the loss by Focal Loss, I am still on the experiment to make sure it works
Not sure why, but this fixes the problem in my case: Firstly, I tried turning on `k4aviewer` and open the device. Then, I set depth mode to Passive IR and...
Thank you for your interest. I wrote some code over SlowFast to extract features for ActivityNet and THUMOS-14. Because I didn't have time to clean the feature extraction code so...
For the videos in Activitynet, I simply rescale them to 1600 frames and extract features with a window size of 16 frames so that each video will be represented by...
> Thank you for your answer. Is the sliding window non overlapping for handling thumos14? Sorry, I think my last comment gets misrendered. I use a sliding window of 128...
Sorry for the late response, please check README for our new linke to object features.
Hi, As annotations on test sub-set is not made publicly available by the organizer of ActivityNet-1.3, we could not do evaluation on testing set. Instead, we use our model to...
can you try below command in the repo ? `cd apex && pip install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./`