Hyunjun Eun

icon indicating a website link https://sites.google.com/view/hyunjun-eun?pli=1

icon location Currently research on Vision-Language Models (VLMs)

Results 5 issues of Hyunjun Eun

feat. length of single segment (16 frames) are 2048, not 512

can you explain about json file? I got 2048 length of the feature for single segment.

can I extract features based on kinetics-64f model?

the outputs of proposal network (probabilities of proposals) are same and very low

1 comment

the outputs of proposal network (probabilities of proposals) are same and very low such as 2.48e-5. So, threshold 0.7 filters all proposals out.

could you release the code to extract optical flow?

2 comment

do I need to normalize RGB images to [0,1] from [0,255] before calculate optical flow? I will follow the code here. https://github.com/opencv/opencv/blob/master/samples/gpu/optical_flow.cpp And 'd_flow' is only output of TVL1. how...

compile with opencv 3.4

I tried to compile opencv 3.0 and 3.1 but failed due to some reasons. I only can compiled opencv 3.4. Thus, any soultion for using C3D with opencv 3.4?