<div>I've been busy the previous week as I had to finish writing my bachelor's thesis on the gestures.</div>
<div>Daniel Willmann - my mentor - sent the Neo FreeRunner to me on Friday. It still hasn't arrived, but it's now in Hungary and will arrive this week. That's definitely good news :)</div>
<div>Like I've said to Mickey and Daniel while we were in the proposal phase of GSoC, I have to implement some tweaks in HMMs to achieve better results.</div>
<div>State duration (a.k.a. duration modeling) will be used to prune unpromising models in the recognizer.</div>
<div> </div>
<div>Like I've said, each model (gesture) has a number of states; each can have its own number of states and these states differ from one model to the other.</div>
<div>Since I'm using a left-to-right model topology, the decoded sequence must be something like this: 1 2 3</div>
<div>where 1 is the initial state, and 3 is the final state.</div>
<div>If the last state is not the final state (the model doesn't reach the final state) then it's clearly not a match (trivial).</div>
<div> </div>
<div>Also here's what happens with models that aren't a match. Note that the length of the state sequence is the length of the observation sequence (HMM assumption).</div>
<div>12333333333333333333333333333333333333333333344</div>
<div>It's clear that the state duration should be proportional with the length and a model that generates a state sequence like that should be pruned.</div>
<div>Instead, I'm looking for something like</div>
<div>1111111122222222223333333333444444444</div>
<div> </div>
<div>This is just a simple optimization that can improve the accuracy. After I prune those models, I look at the posterior probabilities and decide on the best match.</div>
<div> </div>
<div>Paul<br>-- <br><a href="http://www.borza.ro">http://www.borza.ro</a> </div>