Visualizing projected 2D keypoints on the original image

[ericwang0701]

Thank you for sharing your amazing work. I have tried to visualize the projected 2d keypoints (i.e. pj2d) on the original image.
Taking the 3dpw_sit_on_street.jpg sample image as an example. Since pj2d are normalized (-1~1) coordinates on the input image size (512x 512), I first compute their corresponding image coordinate in the original image size (1920x1080) by:
pj2d= (pj2d+ [1,1]) / 2 * [1920, 1080]

Then I plotted those keypoints on the original image. However, it seems there is an offset on the y-axis (i.e. height). All the keypoints are lower than the positions where they should be.

Any thoughts on how to compute and visualize projected 2D keypoints correctly would be really appreciated.

[Arthur151]

The bug could be easily fixed by replace
pj2d= (pj2d+ [1,1]) / 2 * [1920, 1080]
with
pj2d= (pj2d+ [1,1]) / 2 * [1920, 1920]

The reason is that the input image is a square.
Thanks for your interest! Best.

[dae-sun]

Hello, I had same problem but I solved it.
here is solution code.

stride = max(H,W)
pj2d_results[batch_idx]= (pj2d_results[batch_idx]+1.0)/2*stride
j = pj2d_results[batch_idx][0]
img = cv2.circle(img,(int(j[0]),(int(j[1]-(W-H)/2))),10,(0,0,255),-1

[dae-sun]

The bug could be easily fixed by replace
pj2d= (pj2d+ [1,1]) / 2 * [1920, 1080]
with
pj2d= (pj2d+ [1,1]) / 2 * [1920, 1920]

The reason is that the input image is a square.
Thanks for your interest! Best.

You should fix it!
thank you for your code.

[Arthur151]

Thanks for your contribution! I will fix it in the next commit.