Abstraction of Man-Made Shapes
Ravish Mehra, Qingnan Zhou, Jeremy Long, Alla Sheffer, Amy Gooch, Niloy J. Mitra


Man-made objects are ubiquitous in the real world and in virtual environments. While such objects can be very detailed, capturing every small feature, they are often identified and characterized by a small set of defining curves. Compact, abstracted shape descriptions based on such curves are often visually more appealing than the original models, which can appear to be visually cluttered. We introduce a novel algorithm for abstracting three-dimensional geometric models using characteristic curves or contours as building blocks for the abstraction. Our method robustly handles models with poor connectivity, including the extreme cases of polygon soups, common in models of man-made objects taken from online repositories. In our algorithm, we use a two-step procedure that first approximates the input model using a manifold, closed envelope surface and then extracts from it a hierarchical abstraction curve network along with suitable normal information. The constructed curve networks form a compact, yet powerful, representation for the input shapes, retaining their key shape characteristics while discarding minor details and irregularities.


(Left) Vectorization stages: VSA segmentation, segmentation after boundary improvement, smooth approximation geometry, extracted regularized curve network, surface after hierarchical simplification, regularized simplified curve network.
(Right) Input model, processed segments, vectorized curve network, and reconstructed abstraction. Zoom panels show section of curve network, and normals along the curves. Connectivity-only or virtual edges are marked in brown. For ease of visualization, normals from same curve loops are marked in identical colors.
(Below) Result gallery showing various input models, extracted curve networks (with normals), and reconstructed abstractions. The high-resolution abstractions are rendered in yellow, while the low-resolution ones are in blue.
(Right) Abstractions of a microscope model at high- and lowresolutions. For such models with fine details, the abstraction results are often subtle, while geometrically minor but semantically significant parts can get suppressed by our purely geometric abstraction procedure. shadow images.
(Below-left) Fine topological features are easily combined by our envelope construction stage. However, once such features are extracted by a finer grid resolution, we have no easy method to remove them, independent of their size. (Below-right) Some objects, perhaps those less familiar to us, have no obvious natural abstraction.


AUTHOR = "Ravish Mehra and Qingnan Zhou and Jeremy Long and Alla Sheffer and Amy Gooch and Niloy J. Mitra",
TITLE = "Abstraction of Man-Made Shapes",
JOURNAL = "{ACM} Transactions on Graphics",
VOLUME = "28",
NUMBER = "5",
PAGES = "\#137, 1-10",
YEAR = "2009",

paper (30MB) paper (3.5MB) slides (11MB) video data-set/results (160MB)
back to publications
back to homepage