Browsing by Author "Fisher, Matthew"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    Automatic Hierarchical Arrangement of Vector Designs
    (The Eurographics Association, 2021) Fisher, Matthew; Agarwal, Vineet; Beri, Tarun; Theisel, Holger and Wimmer, Michael
    We present a method that transforms an unstructured vector design into a logical hierarchy of groups of objects. Each group is a meaningful collection, formed by proximity in visual characteristics (like size, shape, color, etc.) and spatial location of objects and models the grouping principles designers use. We first simplify the input design by partially or completely flattening it and isolate duplicate geometries in the design (for example, repeating patterns due to copy and paste operations). Next we build the object containment hierarchy by assigning objects that are wholly enclosed inside the geometry of other objects as children of the enclosing parent. In the final clustering phase, we use agglomerative clustering to obtain a bottom-up hierarchical grouping of all objects by comparing and ranking all pairs of objects according to visual and spatial characteristics. Spatial proximity segregates far apart objects, but when they are identical (or near identical) designers generally prefer to keep (and edit) them together. To accommodate this, we detect near identical objects and group them together during clustering despite their spatial separation. We further restrict group formation so that z-order disturbances in the design keep the visual appearance unaffected for tightly-overlapping geometry. The generated organization is equivalent to the original design and the organization results are used to facilitate abstract navigation (hierarchical, lateral or near similar) and selections in the design. Our technique works well with a variety of input designs with commonly identifiable objects and structural patterns.
  • Thumbnail Image
    Item
    Exploratory Stage Lighting Design using Visual Objectives
    (The Eurographics Association and John Wiley & Sons Ltd., 2019) Shimizu, Evan; Paris, Sylvain; Fisher, Matthew; Yumer, Ersin; Fatahalian, Kayvon; Alliez, Pierre and Pellacini, Fabio
    Lighting is a critical element of theater. A lighting designer is responsible for drawing the audience's attention to a specific part of the stage, setting time of day, creating a mood, and conveying emotions. Designers often begin the lighting design process by collecting reference visual imagery that captures different aspects of their artistic intent. Then, they experiment with various lighting options to determine which ideas work best on stage. However, modern stages contain tens to hundreds of lights, and setting each light source's parameters individually to realize an idea is both tedious and requires expert skill. In this paper, we describe an exploratory lighting design tool based on feedback from professional designers. The system extracts abstract visual objectives from reference imagery and applies them to target regions of the stage. Our system can rapidly generate plausible design candidates that embody the visual objectives through a Gibbs sampling method, and present them as a design gallery for rapid exploration and iterative refinement. We demonstrate that the resulting system allows lighting designers of all skill levels to quickly create and communicate complex designs, even for scenes containing many color-changing lights.
  • Thumbnail Image
    Item
    Unsupervised Cycle-consistent Deformation for Shape Matching
    (The Eurographics Association and John Wiley & Sons Ltd., 2019) Groueix, Thibault; Fisher, Matthew; Kim, Vladimir G.; Russel, Bryan C.; Aubry, Mathieu; Bommes, David and Huang, Hui
    We propose a self-supervised approach to deep surface deformation. Given a pair of shapes, our algorithm directly predicts a parametric transformation from one shape to the other respecting correspondences. Our insight is to use cycle-consistency to define a notion of good correspondences in groups of objects and use it as a supervisory signal to train our network. Our method combines does not rely on a template, assume near isometric deformations or rely on point-correspondence supervision. We demonstrate the efficacy of our approach by using it to transfer segmentation across shapes. We show, on Shapenet, that our approach is competitive with comparable state-of-the-art methods when annotated training data is readily available, but outperforms them by a large margin in the few-shot segmentation scenario.