Search for: All records

We introduce ShapeCoder, the first system capable of taking a dataset of shapes, represented with unstructured primitives, and jointly discovering (i) usefulabstractionfunctions and (ii) programs that use these abstractions to explain the input shapes. The discovered abstractions capture common patterns (both structural and parametric) across a dataset, so that programs rewritten with these abstractions are more compact, and suppress spurious degrees of freedom. ShapeCoder improves upon previous abstraction discovery methods, finding better abstractions, for more complex inputs, under less stringent input assumptions. This is principally made possible by two methodological advancements: (a) a shape-to-program recognition network that learns to solve sub-problems and (b) the use of e-graphs, augmented with a conditional rewrite scheme, to determine when abstractions with complex parametric expressions can be applied, in a tractable manner. We evaluate ShapeCoder on multiple datasets of 3D shapes, where primitive decompositions are either parsed from manual annotations or produced by an unsupervised cuboid abstraction method. In all domains, ShapeCoder discovers a library of abstractions that captures high-level relationships, removes extraneous degrees of freedom, and achieves better dataset compression compared with alternative approaches. Finally, we investigate how programs rewritten to use discovered abstractions prove useful for downstream tasks.

 

Advanced sequencing technologies have expedited resolution of higher-level arthropod relationships. Yet, dark branches persist, principally among groups occurring in cryptic habitats. Among chelicerates, Solifugae ("camel spiders") is the last order lacking a higher-level phylogeny and have thus been historically characterized as "neglected [arachnid] cousins". Though renowned for aggression, remarkable running speed, and xeric adaptation, inferring solifuge relationships has been hindered by inaccessibility of diagnostic morphological characters, whereas molecular investigations have been limited to one of 12 recognized families. Our phylogenomic dataset via capture of ultraconserved elements sampling all extant families recovered a well-resolved phylogeny, with two distinct groups of New World taxa nested within a broader Paleotropical radiation. Divergence times using fossil calibrations inferred that Solifugae radiated by the Permian, and most families diverged prior to the Paleogene-Cretaceous extinction, likely driven by continental breakup. We establish Boreosolifugae new suborder uniting five Laurasian families, and Australosolifugae new suborder uniting seven Gondwanan families using morphological and biogeographic signal. 

We present SHRED, a method for 3D SHape REgion Decomposition. SHRED takes a 3D point cloud as input and uses learned local operations to produce a segmentation that approximates fine-grained part instances. We endow SHRED with three decomposition operations: splitting regions, fixing the boundaries between regions, and merging regions together. Modules are trained independently and locally, allowing SHRED to generate high-quality segmentations for categories not seen during training. We train and evaluate SHRED with fine-grained segmentations from PartNet; using its merge-threshold hyperparameter, we show that SHRED produces segmentations that better respect ground-truth annotations compared with baseline methods, at any desired decomposition granularity. Finally, we demonstrate that SHRED is useful for downstream applications, out-performing all baselines on zero-shot fine-grained part instance segmentation and few-shot finegrained semantic segmentation when combined with methods that learn to label shape regions. 

Procedural models (i.e. symbolic programs that output visual data) are a historically‐popular method for representing graphics content: vegetation, buildings, textures, etc. They offer many advantages: interpretable design parameters, stochastic variations, high‐quality outputs, compact representation, and more. But they also have some limitations, such as the difficulty of authoring a procedural model from scratch. More recently, AI‐based methods, and especially neural networks, have become popular for creating graphic content. These techniques allow users to directly specify desired properties of the artifact they want to create (via examples, constraints, or objectives), while a search, optimization, or learning algorithm takes care of the details. However, this ease of use comes at a cost, as it's often hard to interpret or manipulate these representations. In this state‐of‐the‐art report, we summarize research on neurosymbolic models in computer graphics: methods that combine the strengths of both AI and symbolic programs to represent, generate, and manipulate visual data. We survey recent work applying these techniques to represent 2D shapes, 3D shapes, and materials & textures. Along the way, we situate each prior work in a unified design space for neurosymbolic models, which helps reveal underexplored areas and opportunities for future research.

 

In this Perspective, we summarize the status of technological development for large-area and low-noise substrate-transferred GaAs/AlGaAs (AlGaAs) crystalline coatings for interferometric gravitational-wave (GW) detectors. These topics were originally presented as part of an AlGaAs Workshop held at American University, Washington, DC, from 15 August to 17 August 2022, bringing together members of the GW community from the laser interferometer gravitational-wave observatory (LIGO), Virgo, and KAGRA collaborations, along with scientists from the precision optical metrology community, and industry partners with extensive expertise in the manufacturing of said coatings. AlGaAs-based crystalline coatings present the possibility of GW observatories having significantly greater range than current systems employing ion-beam sputtered mirrors. Given the low thermal noise of AlGaAs at room temperature, GW detectors could realize these significant sensitivity gains while potentially avoiding cryogenic operation. However, the development of large-area AlGaAs coatings presents unique challenges. Herein, we describe recent research and development efforts relevant to crystalline coatings, covering characterization efforts on novel noise processes as well as optical metrology on large-area (∼10 cm diameter) mirrors. We further explore options to expand the maximum coating diameter to 20 cm and beyond, forging a path to produce low-noise mirrors amenable to future GW detector upgrades, while noting the unique requirements and prospective experimental testbeds for these semiconductor-based coatings.