3D modeling of multiple-object scenes from sets of images

Grum, Matthew; Borş, Adrian G.

doi:10.1016/j.patcog.2013.04.020

Cited by 10 publications

(1 citation statement)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This method accelerates the computation and increases the representation accuracy. With the introduction of Visual Hull (VH) concept Image Processing Theory, Tools and Applications 978-1-4799-8637-8/15/$31.00 ©2015 IEEE [5], the silhouette-based 3D reconstruction research has attracted lots of attention [8], [9], [10], [11], [12]. Kutulakos and Seitz [4] define a unified framework to construct a 3D model from multi-view silhouettes.…”

Section: Related Workmentioning

confidence: 99%

Silhouette-based 3D model for zebrafish high-throughput imaging

Guo

Veneman

Spaink

et al. 2015

2015 International Conference on Image Processing Theory, Tools and Applications (IPTA)

View full text Add to dashboard Cite

Good estimations of volume and surface area are important to biological systems measurement. In this paper we develop a 3D reconstruction from evenly sampled axial views in order to enable the volume and surface area measurement. We develop this system for high throughput applications with the zebrafish model system. The VAST BioImager is specifically developed for this purpose and with this system the axial views can be produced. Silhouettes derived from the axial sequence are shape priors which can be directly used to solve the camera calibration problem that is required for the accurate 3D reconstruction. Nonlinear optimisation algorithms have shown to be suitable for the further development of the reconstruction problem. The method proposed in this paper can be included in a measurement pipeline that is used in all kinds of high throughput applications in the zebrafish field. From the 3D reconstruction features can be derived that will contribute to the phenotyping of zebrafish.

show abstract