Closed W4ngH4o closed 2 years ago
I'm not sure what problem you met. For your question, since the visualization is forked from PySlowfast, I suggest you open an issue in their repo.
For my visualization, I use this repo. Since it is designed for images, for videos, I use grad_cam in each frame, and interpolate the heatmap if the frames are downsampled.
I'm not sure what problem you met. For your question, since the visualization is forked from PySlowfast, I suggest you open an issue in their repo.
For my visualization, I use this repo. Since it is designed for images, for videos, I use grad_cam in each frame, and interpolate the heatmap if the frames are downsampled.
Thanks for your suggestion. I have resolved my problem.
As there is no more activity, I am closing the issue, don't hesitate to reopen it if necessary.
When I utilize Grad_cam to visualize, the localization_map is not all 0. But after plt.get_cmap, the heatmap is all 0, which leads the visualization is black. I am confused and looking forward to your reply. Thanks so much.