TY - GEN
T1 - Score-based diffusion priors for multi-target detection
AU - Zabatani, Alon
AU - Kreymer, Shay
AU - Bendory, Tamir
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - Multi-target detection (MTD) is the problem of estimating an image from a large, noisy measurement that contains randomly translated and rotated copies of the image. Motivated by the single-particle cryo-electron microscopy technology, we design data-driven diffusion priors for the MTD problem, derived from score-based stochastic differential equations models. We then integrate the prior into the approximate expectation-maximization algorithm. In particular, our method alternates between an expectation step that approximates the expected log-likelihood and a maximization step that balances the approximated log-likelihood with the learned log-prior. We show on two datasets that adding the data-driven prior substantially reduces the estimation error, in particular in high noise regimes.
AB - Multi-target detection (MTD) is the problem of estimating an image from a large, noisy measurement that contains randomly translated and rotated copies of the image. Motivated by the single-particle cryo-electron microscopy technology, we design data-driven diffusion priors for the MTD problem, derived from score-based stochastic differential equations models. We then integrate the prior into the approximate expectation-maximization algorithm. In particular, our method alternates between an expectation step that approximates the expected log-likelihood and a maximization step that balances the approximated log-likelihood with the learned log-prior. We show on two datasets that adding the data-driven prior substantially reduces the estimation error, in particular in high noise regimes.
KW - Diffusion models
KW - cryo-EM
KW - expectation-maximization
KW - multi-target detection
KW - score-SDE
UR - http://www.scopus.com/inward/record.url?scp=85190626444&partnerID=8YFLogxK
U2 - 10.1109/CISS59072.2024.10480190
DO - 10.1109/CISS59072.2024.10480190
M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???
AN - SCOPUS:85190626444
T3 - 2024 58th Annual Conference on Information Sciences and Systems, CISS 2024
BT - 2024 58th Annual Conference on Information Sciences and Systems, CISS 2024
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 58th Annual Conference on Information Sciences and Systems, CISS 2024
Y2 - 13 March 2024 through 15 March 2024
ER -