We propose a novel algorithm for decomposing general three-dimensional geometries into a small set of overlap-free height-field blocks , volumes enclosed by a flat base and a height-field surface defined with respect to this base. This decomposition is useful for fabrication methodologies such as 3-axis CNC milling, where a single milling pass can only carve a single height-field surface defined with respect to the machine tray but can also benefit other fabrication settings. Computing our desired decomposition requires solving a highly constrained discrete optimization problem, variants of which are known to be NP-hard. We effectively compute a high-quality decomposition by using a two-step process that leverages the unique characteristics of our setup. Specifically, we notice that if the height-field directions are constrained to the major axes, then we can always produce a valid decomposition starting from a suitable surface segmentation. Our method first produces a compact set of large, possibly overlapping, height-field blocks that jointly cover the model surface by recasting this discrete constrained optimization problem as an unconstrained optimization of a continuous function, which allows for an efficient solution. We then cast the computation of an overlap-free, final decomposition as an ordering problem on a graph and solve it via a combination of cycle elimination and topological sorting. The combined algorithm produces a compact set of height-field blocks that jointly describe the input model within a user given tolerance. We demonstrate our method on a range of inputs and showcase a number of real life models manufactured using our technique.

We propose a novel method for the automatic generation of structured hexahedral meshes of articulated 3D shapes. We recast the complex problem of generating the connectivity of a hexahedral mesh of a general shape into the simpler problem of generating the connectivity of a tubular structure derived from its curve-skeleton. We also provide volumetric subdivision schemes to nicely adapt the topology of the mesh to the local thickness of tubes, while regularizing per-element size. Our method is fast, one-click, easy to reproduce, and it generates structured meshes that better align to the branching structure of the input shape if compared to previous methods for hexa mesh generation

Semantic segmentation is a widespread image analysis task; in some applications, it requires such high accuracy that it still has to be done manually, taking a long time. Deep learning‐based approaches can significantly reduce such times, but current automated solutions may produce results below expert standards. We propose agLab, an interactive tool for the rapid labelling and analysis of orthoimages that speeds up semantic segmentation. TagLab follows a human‐centered artificial intelligence approach that, by integrating multiple degrees of automation, empowers human capabilities. We evaluated TagLab's efficiency in annotation time and accuracy through a user study based on a highly challenging task: the semantic segmentation of coral communities in marine ecology. In the assisted labelling of corals, TagLab increased the annotation speed by approximately 90% for nonexpert annotators while preserving the labelling accuracy. Furthermore, human–machine interaction has improved the accuracy of fully automatic predictions by about 7% on average and by 14% when the model generalizes poorly. Considering the experience done through the user study, TagLab has been improved, and preliminary investigations suggest a further significant reduction in annotation times.

Figure 1: Our processing pipeline in a nutshell (left to right): we first compute the rotation axis on the input mesh; we then partition the mesh in millable height-field portions (including the top and bottom ones) taking also into account the information of the saliency map; we compute a milling sequence and fabricate the object using the 4-axis milling machine; we clean up the result to obtain the final real object.

A B S T R A C TIn recent years, fabrication technologies have developed at a breakneck pace. However, some limitations on shape and dimension still apply both to additive and subtractive manufacturing, and one way to bypass them could be the partition of the object to build. We present here a novel algorithm, based on the polycube representation of the original shape, able to decompose any model into smaller parts simpler to fabricate. We first map the shape in a polycube and, then, split it to take advantage of the polycube partitioning. In this way, we obtain quite easily a partition of the model. In this work we also study and analyze pros and cons of this partitioning scheme for fabrication, when using both the additive and subtractive pipelines. Our proposed partitioning scheme is computationally light, and it produces high-grade results, especially when applied to models that we can map onto polycubes with a high compactness value.