You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Many thanks for providing the code...
If I would like to use other resolution like 512x512 or 1024x512,
Could I only change the dim parameter in deploy.txt ?
I am trying to only change the input_shape and test python code like following,
input_shape {
dim: 1
dim: 3
The inference result is bad, some images will get multiple bounding boxes.
Do you have any suggestion if the input resolution is 512x512 or 1024x512?
Do I need to retain this input resolution?
Thanks again,
Jerry Ho
The text was updated successfully, but these errors were encountered:
jerryho-quanta
changed the title
First of all many thanks for providing the code, greatly appreciated!!
Pelee input resolution problem
Feb 25, 2021
Many thanks for providing the code...
If I would like to use other resolution like 512x512 or 1024x512,
Could I only change the dim parameter in deploy.txt ?
I am trying to only change the input_shape and test python code like following,
input_shape {
dim: 1
dim: 3
dim: 304
dim: 304
dim: 512
dim: 512
}
set net to batch size of 1
#image_resize_w = 304
image_resize_w = 512
image_resize_h = 512
net.blobs['data'].reshape(1,3,image_resize_w,image_resize_h)
The inference result is bad, some images will get multiple bounding boxes.
Do you have any suggestion if the input resolution is 512x512 or 1024x512?
Do I need to retain this input resolution?
Thanks again,
Jerry Ho
The text was updated successfully, but these errors were encountered: