search

mks0601 / I2L-MeshNet_RELEASE

Official PyTorch implementation of "I2L-MeshNet: Image-to-Lixel Prediction Network for Accurate 3D Human Pose and Mesh Estimation from a Single RGB Image", ECCV 2020
MIT License
710 stars 130 forks source link

Some bounding box coordinates negative #48

Closed Shubhendu-Jena closed 3 years ago

Shubhendu-Jena commented 3 years ago

Hi,

Thank you for the great work. I had one doubt. Some of the bounding box coordinates that I am getting as output from "process_bbox" function are negative. Is this deliberate? One example of such an instance is for the image "s_05_act_04_subact_01_ca_04_003112.jpg". For this image, the bounding box origin coordinates, width and height are : array([ -6.87618172, 84.10320063, 606.78921781, 606.78921781]). I found this out because I was trying to save the cropped images according to the bounding box coordinates.

Looking forward to your response

mks0601 commented 3 years ago

Hi, as I'm using affine transformation to crop and resize the human image from the original image (see here), the negative coordinates are not problems.

Shubhendu-Jena commented 3 years ago

Hi, thanks for the reply. Yes, it won't cause any problems regarding the re-sizing of course. I just wanted to know why the coordinates are negative. I was trying to save the cropped images inside the bounding box, hence I was facing this issue.

mks0601 commented 3 years ago

Could you see the function at here? It re-calculates coordinates based on scaled widths and heights.

Shubhendu-Jena commented 3 years ago

Hi, thanks again for the response. Got it. Closing issue.