Markov Models

This video is part of the Udacity course "Introduction to Computer Vision". Watch the full course at www.udacity.co...

[@SilvioAnkermann]

Good explanation! Just one thing: At 2:20 it's the other way round. First goes the future state, then comes the current state. So a_(ij) = P(q_(t+1) = S_j | q_t = S_i)

[@thomaspeterson2568]

Yes, I was thinking the same!

[@nonoleg]

Same thought, i'm not crazy ! :)

[@ersindeger5155]

Thanks for pointing out ... İ was going crazy with rows vs columns for a few minutes

[@darshitthakar7999]

Well pointed

[@artinbogdanov]

+1

[@suradegefu1664]

And also your lecture is great!

[@kabirkanha]

Oh my god, that was amazing

[@Nimolo96]

What is the address for the post card? :D

[@nivu]

Wow, Now that's a Markov Model :)

[@basiccstutorials8346]

Awsome explanation

[@krishnachaitanyajagadeeshs712]

Great Video.
Thumbs up..

[@rhetorical_annihilation]

Nice explanation.

[@sambitmukherjee1713]

Superb explanation...

[@kavyadanivas9894]

Really nice. can you explain HMM in speech recognition please. Thank you

[@VeronaLynne]

Where'd you get the initial state distribution from?? (3:06)

[@kush5749]

it mostly depends on today. but he assumed those values

[@0xAob]

็How to collect the data for probability transition state and initial state in the real life

[@panahi3446]

how did you calculate the Initial state distribution?

[@kush5749]

it provided i guess

[@shashankesh]

u got me

[@AdeelKhalidSE]

I don't understand the Initial state distribution part, how you calculated that one? If you chose snowy as the first state then there are 75% chances of sunny, 5 % chances for rainy and 20% of snowy, correct me if I'm wrong.

[@lospolloshermanos6213]

@Konstantinos Mechteridis this could be based on a prior

[@quannmtt3110]

Awsome explanation