Teaser: Given the 3D buddha head model (a), the object printed using our slicing algorithm (b) has indistinguishable visual quality with the finest one printed by the highest resolution and saves 29.44% printing time. By applying the segmentation method, the segmented object (c) is printed which saves 39.11% printing time while also preserves the visual appearance (d). All of the savings are relative to the objects printed with finest resolution.
Abstract
We present an adaptive slicing scheme for reducing manufacturing time of
3D printing system. Based on a new saliency based metric, our method
optimizes the thicknesses of slicing layers to save the printing time
and preserve the visual quality of printing results. We formulate the
problem as a constrained l0 optimization and compute the slicing result
via a two-step optimization scheme. To further reduce the printing time,
we develop a saliency based segmentation scheme to partition an object
into subparts and then optimize the slicing of each subpart separately.
We validate our method with a large set of 3D shapes ranging from CAD
models to scanned objects. Results show that our method saves 30-40%
printing time and generates 3D objects that are visually similar to the
ones printed with the finest resolution.
Figure 1. Overview of our adaptive slicing algorithm. Given an input model (a), our algorithm initializes the slicing result (b). Then our algorithm iteratively executes timing optimization (c) and visual optimization (d) until the visual degradation of the sliced mesh exceeds the user specified threshold. The final slicing result is (e).
Figure 2. Comparison of different slicing results for the Buddha head model. The first row shows the close-up photos of the regions in red and the second row shows the photos of printed objects.
Figure 3. Photos of the printed objects using AdapSlice. Numbers in brackets below each photo denote its printing time (in minutes) and its ratio of time saving according to the ground truth. As shown in the user study, there is no apparent visual difference between these objects and their corresponding ground truth.
We would like to thank the reviewers for their detailed comments and suggestions which greatly improved the manuscript. We are greatful to Lian et al. [LGB*11] and Shilane et al. [SMKF04] for their 3D databases. The work is supported by the One Hundred Talent Project of the Chinese Academy of Sciences and Natural Science Foundation of China (61222206, 61202284, 61173102, 61370143).
title = {Saliency-Preserving Slicing Optimization for Effective 3D Printing},
author = {Weiming Wang and Haiyuan Chao and Jing Tong and Zhouwang Yang and Xin Tong and Hang Li and Xiuping Liu and Ligang Liu}
journal = {Computer Graphics Forum},
volume={},
number={},
pages={},
year = {2015}
}