Devarajan Ramanujan, William Benjamin, William Z. Benstein, Niklas Elmqvist, Karthik Ramani
shapeSift: Suggesting sustainable options in design reuse from part repositories

shapeSift is a framework for supporting sustainability based decision making during selection of similar previous designs from part repositories. Our framework is designed for 3D part repositories that contain metadata pertaining to materials and manufacturing processes as well as the functionality of a given part. To demonstrate the usefulness of the proposed framework we develop an example multi-dimensional visualization that encodes part similarities as well as a calculated environmental sustainability indicator. This visualization is incorporated in a prototype interface that focuses on enhancing the intuitiveness of the exploration process by the use of sketch-based retrieval.


Visualization pipeline for exploring 3d repositories using the shapeSift framework. Users begin by sketch based querying as shown in (1). Our system organizes search results using a squarified layout (2) that is constructed using Brul’s squarified treemap algorithm. Here, each color corresponds to a particular material class. The area of a cell is scaled in negative proportion to the calculated environmental impact indicator. Parts are ordered by shape similarity relative to the query. Users can explore these results further using a similarity polygon (3a), sliders that set values for distance threshold (3b) and interactive tooltips (3c). The combination of these steps forms a unit iteration that can be repeated as desired by the user.

Demonstration Video


Posted in 2013, Design, Design and Learning, Devarajan Ramanujan, Karthik Ramani, Niklas Elmqvist, William Benjamin, William Z Bernstein