Dynamical Statistical Shape Priors for Level Set-Based Tracking
IEEE Transactions on Pattern Analysis and Machine Intelligence
Motion-based enhancement of optical imaging
ISBI'09 Proceedings of the Sixth IEEE international conference on Symposium on Biomedical Imaging: From Nano to Macro
Diffeomorphic registration using b-splines
MICCAI'06 Proceedings of the 9th international conference on Medical Image Computing and Computer-Assisted Intervention - Volume Part II
| Hi-index | 0.00 |
Optical imaging is an efficient mean to measure biological signal. However, it can suffer from low spatial and temporal resolution while animal deformable displacements could also degrade significantly the localization of the measurements. In this paper, we propose a novel approach to perform fusion of cinematic flow and optical imaging towards enhancement of the biological signal. To this end, fusion is reformulated as a population (all vs. all) registration problem where the two (being spatially aligned) signals are registered in time using the same deformation field. Implicit silhouette and landmark matching are considered for the cinematic images and are combined with global statistical congealing-type measurements of the optical one. The problem is reformulated using a discrete MRF, where optical imaging costs are expressed in singleton (global) potentials, while smoothness constraints as well as cinematic measurements through pair-wise potentials. Promising experimental results demonstrate the potentials of our approach.