lzccccc / 3d-bounding-box-estimation-for-autonomous-driving

3d bounding box estimation from monocular image based on 2d bounding box
127 stars 36 forks source link

how to plot 3dbb without P2 #10

Closed yasserius closed 4 years ago

yasserius commented 4 years ago

Hi,

firstly, excellent repo, very nicely written code, love it!

But I am confused a bit with the math.

The model outputs 2 orientation values, a sin and a cosine of the local orientation alpha, as predictions.

But while looking at visualization3Dbox.py, it appears that it used the P2 provided by the calib data.

So how will plotting happen for an unseen image, when P2 for it isn't available?

Let me know if I failed to explain myself. Thanks!

lzccccc commented 4 years ago

Hi,

It is not possible if you don't know P2. This is called the camera intrinsic matrix which transforms the camera coordinate to pixel coordinate.

yasserius commented 4 years ago

That's a bummer.

So it means if I have some random footage of the road, then this model doesn't work?

If so, are there any other models that you know that are independent of P2?