fighting!

when I use your detectron ,I find my mask_loss didn't decrease? what happen ,and why?

first of all, thanks for your great job. To be honest, I am very confused about ortho_ratio what is the meaning of ortho_ratio, I see your code in render_data.py 2/(ortho_ratio*img_size)...

how tow concat roi feature with pose

1

I am very interested in your work. In your paper, you said concate Roi feature with pose. Do you compress the feature to d-dimension? My understanding of your approach is：...

I watch your code in evaluate. I find your NMS process all proposals bbox. It's make some trouble in early epoch due to some many remained proposals. So, I Suggest...

When I use your refinedet to train our data ,I find the loss is very big why ?

3

for example the train loss for total_loss as to 4.0 ? why this condition happen? do you divide batch_size ? please help me

Focal loss

2

我感觉你的focal loss 有点和论文不一样是你自己修改的还是,这么样的呢?可以给我些解答吗

Thank your for your good job. We follow your work, but we only find coco weight for 17 joints. could you give our your crowdpose weight based on pytorch? Thank...

Do you have a segmentation part?

1

In the paper you said using NMS to get the identifiers for each object instance but I don't know how to identifiers! We calculate a histogram of the tags and...

we have some trouble . when using your code ,for make ,how to make it ? the OpenCV_INCLUDE_DIRS and OpenCV_LIBS how to use

error: command 'x86_64-linux-gnu-gcc' failed with exit status 1

1

There is a question mentioned above. My torch vision is 1.6 whether is higher than your version?