Seanforfun / GMAN_Net_Haze_Removal

Single Image Dehazing with a Generic Model-Agnostic Convolutional Neural Network

https://ieeexplore.ieee.org/document/8686264

38 stars 21 forks source link

Alpha Estimation Statistical Result #5

Closed Seanforfun closed 6 years ago

Seanforfun commented 6 years ago

Alpha Estiamtion

Please click each error rate to get details!

Highest Intensity: Sort the average value of RGB channels and take the highest as result.
Smallest Channel Difference: value = (Vr - Vg) ^2 + (Vr - Vb) ^2 + (Vb - Vg) ^2, take the minimun.
Highest Channel Value: Pick the highest value of all three channels of those selected pixels, in another word, pick the largest value.

Indoor	DCP(DarkChannelTransmissionMap)	Attenuation(DepthMap)	Deep Learning Method
Highest Intensity	8.096%	7.831%	TBD
Smallest Channel Difference	6.748%	7.546%	TBD
Highest Channel Value	7.638%	7.189%	TBD

Seanforfun commented 6 years ago

Alpha Estiamtion

Outdoor	DCP(DarkChannelTransmissionMap)	Attenuation(DepthMap)	Deep Learning Method
Highest Intensity	7.670%	7.674%	TBD
Smallest Channel Difference	7.569%	7.634%	TBD
Highest Channel Value	7.969%	8.029%	TBD