Trjtdtk

Why does processed meat contain preservatives, while canned fish needs not?

What does it mean to express a gate in Dirac notation?

To say I met a person for the first time

Combinable filters

How do I deal with a coworker that keeps asking to make small superficial changes to a report, and it is seriously triggering my anxiety?

What's the polite way to say "I need to urinate"?

Don’t seats that recline flat defeat the purpose of having seatbelts?

What makes accurate emulation of old systems a difficult task?

Reducing vertical space in stackrel

Apply MapThread to all but one variable

A Strange Latex Symbol

Why is it that the natural deduction method can't test for invalidity?

Error message with tabularx

What does KSP mean?

What route did the Hindenburg take when traveling from Germany to the U.S.?

Controversial area of mathematics

Binary Numbers Magic Trick

How can Republicans who favour free markets, consistently express anger when they don't like the outcome of that choice?

Will a top journal at least read my introduction?

How do I reattach a shelf to the wall when it ripped out of the wall?

Is there a way to get a compiler for the original B programming language?

Size of electromagnet needed to replicate Earth's magnetic field

Realistic Necromancy?

How to creep the reader out with what seems like a normal person?

Bouding box regression in R-CNN (and similar methods)

Input and output feature shapes in CNN for speech recognitionShould the output of regression models, like SVR, be normalized?Preparing ground truth labels for YOLO3Contextual Object DetectionForward or Backward Stepwise RegressionHow to obtain and load a good initial data set for object localization?Training detector without bounding box dataTattoo Image Recognition - Should I Crop Training Data BackgroundMean Absolute Error in Random Forest RegressionBounding box regression in R-CNN

0

$begingroup$

Here is the bounding-box regression technique used in R-CNN

What I don't understand is why $t_w$ needs to be $log (G_w / P_w)$ instead of just $G_w / P_w$ (i.e. using normalized coordinates just like $t_x$ and $t_y$)

Can anyone explain it for me ?

regression object-detection

share|improve this question

asked Apr 7 at 5:59

HOANG GIANGHOANG GIANG

426

$endgroup$

add a comment | 

0

$begingroup$

Here is the bounding-box regression technique used in R-CNN

What I don't understand is why $t_w$ needs to be $log (G_w / P_w)$ instead of just $G_w / P_w$ (i.e. using normalized coordinates just like $t_x$ and $t_y$)

Can anyone explain it for me ?

regression object-detection

share|improve this question

asked Apr 7 at 5:59

HOANG GIANGHOANG GIANG

426

$endgroup$

add a comment | 

$begingroup$

Here is the bounding-box regression technique used in R-CNN

What I don't understand is why $t_w$ needs to be $log (G_w / P_w)$ instead of just $G_w / P_w$ (i.e. using normalized coordinates just like $t_x$ and $t_y$)

Can anyone explain it for me ?

regression object-detection

share|improve this question

asked Apr 7 at 5:59

HOANG GIANGHOANG GIANG

426

$endgroup$

share|improve this question

asked Apr 7 at 5:59

HOANG GIANGHOANG GIANG

426

share|improve this question

asked Apr 7 at 5:59

HOANG GIANGHOANG GIANG

426

asked Apr 7 at 5:59

HOANG GIANGHOANG GIANG

426

asked Apr 7 at 5:59

HOANG GIANGHOANG GIANG

426