HIGHLIGHTS
SUMMARY
In metrology, once an instrument is calibrated, one naturally should quantify the following properties of the model: (a) (b) (c) Consistency: How well does the calibrated model agree with the calibration data? The end user/application then is ultimately interested in the geometry of uncontrolled scenes recorded under similar conditions, e_g, sizes and shapes of objects and/or positions and orientations of the camera itself. The basics of projective geometry may be found, e_g, As an alternative, some recent works demonstrate that multi-layer neural_networks can be trained to infer some camera parameters using . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.