Volumetric bias in segmentation and reconstruction : Secrets and solutions

Many standard optimization methods for segmentation and reconstruction compute ML model estimates for ap- pearance or geometry of segments, e.g. Zhu-Yuille [23], Torr [20], Chan-Vese [6], GrabCut [18], Delong et al. [8]. We observe that the standard likelihood term in these formu- lations corresponds to a generalized probabilistic K-means energy. In learning it is well known that this energy has a strong bias to clusters of equal size [11], which we express as a penalty for KL divergence from a uniform distribution of cardinalities. However, this volumetric bias has been mostly ignored in computer vision. We demonstrate signif- icant artifacts in standard segmentation and reconstruction methods due to this bias. Moreover, we propose binary and multi-label optimization techniques that either (a) remove this bias or (b) replace it by a KL divergence term for any given target volume distribution. Our general ideas apply to continuous or discrete energy formulations in segmenta- tion, stereo, and other reconstruction problems.

Isack, HossamUniversity of Western Ontario (författare)
Olsson, CarlLund University,Lunds universitet,Matematik LTH,Matematikcentrum,Institutioner vid LTH,Lunds Tekniska Högskola,Mathematics (Faculty of Engineering),Centre for Mathematical Sciences,Departments at LTH,Faculty of Engineering, LTH(Swepub:lu)math-coo (författare)
Ayed, Ismail BenÉcole de technologie supérieure ETS (författare)
University of Western OntarioMatematik LTH (creator_code:org_t)

Ingår i:Proceedings - 2015 IEEE International Conference on Computer Vision, ICCV 201511-18-December-2015, s. 1769-17779781467383912

Av författaren/redakt...: Boykov, Yuri; Isack, Hossam; Olsson, Carl; Ayed, Ismail Ben

Om ämnet

NATURVETENSKAP: NATURVETENSKAP; och Data och informa ...; och Datorseende och ...

Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.