?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.title=Expressing+Bayesian+Fusion+as+a+Product+of+Distributions%3A+Application+to+Randomized+Hough+Transform&rft.creator=Pradalier%2C+C&rft.creator=Colas%2C+F&rft.creator=Bessiere%2C+P&rft.subject=Robotics&rft.description=Data+fusion+is+a+common+issue+of+mobile+robotics%2C+computer+assisted%0Amedical+diagnosis+or+behavioral+control+of+simulated+character+for+instance.+However%0Adata+sources+are+often+noisy%2C+opinion+for+experts+are+not+known+with+absolute%0Aprecision%2C+and+motor+commands+do+not+act+in+the+same+exact+manner+on+the+environment.%0AIn+these+cases%2C+classic+logic+fails+to+manage+efficiently+the+fusion+process.%0AConfronting+different+knowledge+in+an+uncertain+environment+can+therefore+be+adequately%0Aformalized+in+the+bayesian+framework.%0ABesides%2C+bayesian+fusion+can+be+expensive+in+terms+of+memory+usage+and+processing%0Atime.+This+paper+precisely+aims+at+expressing+any+bayesian+fusion+process+as+a%0Aproduct+of+probability+distributions+in+order+to+reduce+its+complexity.+We+first+study%0Aboth+direct+and+inverse+fusion+schemes.+We+show+that+contrary+to+direct+models%2C%0Ainverse+local+models+need+a+specific+prior+in+order+to+allow+the+fusion+to+be+computed%0Aas+a+product.+We+therefore+propose+to+add+a+consistency+variable+to+each+local%0Amodel+and+we+show+that+these+additional+variables+allow+the+use+of+a+product+of+the%0Alocal+distributions+in+order+to+compute+the+global+probability+distribution+over+the%0Afused+variable.+Finally%2C+we+take+the+example+of+the+Randomized+Hough+Transform.%0AWe+rewrite+it+in+the+bayesian+framework%2C+considering+that+it+is+a+fusion+process%0Ato+extract+lines+from+couples+of+dots+in+a+picture.+As+expected%2C+we+can+find+back%0Athe+expression+of+the+Randomized+Hough+Transform+from+the+literature+with+the%0Aappropriate+assumptions.&rft.date=2003&rft.type=Conference+Paper&rft.type=NonPeerReviewed&rft.format=application%2Fpdf&rft.identifier=http%3A%2F%2Fcogprints.org%2F3758%2F1%2FPradalier03a.pdf&rft.identifier=++Pradalier%2C+C+and+Colas%2C+F+and+Bessiere%2C+P++(2003)+Expressing+Bayesian+Fusion+as+a+Product+of+Distributions%3A+Application+to+Randomized+Hough+Transform.++%5BConference+Paper%5D+++++&rft.relation=http%3A%2F%2Fcogprints.org%2F3758%2F