-
LSD-StructureNet: Modeling Levels of Structural Detail in 3D Part Hierarchies
Authors:
Dominic Roberts,
Ara Danielyan,
Hang Chu,
Mani Golparvar-Fard,
David Forsyth
Abstract:
Generative models for 3D shapes represented by hierarchies of parts can generate realistic and diverse sets of outputs. However, existing models suffer from the key practical limitation of modelling shapes holistically and thus cannot perform conditional sampling, i.e. they are not able to generate variants on individual parts of generated shapes without modifying the rest of the shape. This is li…
▽ More
Generative models for 3D shapes represented by hierarchies of parts can generate realistic and diverse sets of outputs. However, existing models suffer from the key practical limitation of modelling shapes holistically and thus cannot perform conditional sampling, i.e. they are not able to generate variants on individual parts of generated shapes without modifying the rest of the shape. This is limiting for applications such as 3D CAD design that involve adjusting created shapes at multiple levels of detail. To address this, we introduce LSD-StructureNet, an augmentation to the StructureNet architecture that enables re-generation of parts situated at arbitrary positions in the hierarchies of its outputs. We achieve this by learning individual, probabilistic conditional decoders for each hierarchy depth. We evaluate LSD-StructureNet on the PartNet dataset, the largest dataset of 3D shapes represented by hierarchies of parts. Our results show that contrarily to existing methods, LSD-StructureNet can perform conditional sampling without impacting inference speed or the realism and diversity of its outputs.
△ Less
Submitted 7 September, 2021; v1 submitted 18 August, 2021;
originally announced August 2021.
-
ConstructAide: Analyzing and Visualizing Construction Sites through Photographs and Building Models
Authors:
Kevin Karsch,
Mani Golparvar-Fard,
David Forsyth
Abstract:
We describe a set of tools for analyzing, visualizing, and assessing architectural/construction progress with unordered photo collections and 3D building models. With our interface, a user guides the registration of the model in one of the images, and our system automatically computes the alignment for the rest of the photos using a novel Structure-from-Motion (SfM) technique; images with nearby v…
▽ More
We describe a set of tools for analyzing, visualizing, and assessing architectural/construction progress with unordered photo collections and 3D building models. With our interface, a user guides the registration of the model in one of the images, and our system automatically computes the alignment for the rest of the photos using a novel Structure-from-Motion (SfM) technique; images with nearby viewpoints are also brought into alignment with each other. After aligning the photo(s) and model(s), our system allows a user, such as a project manager or facility owner, to explore the construction site seamlessly in time, monitor the progress of construction, assess errors and deviations, and create photorealistic architectural visualizations. These interactions are facilitated by automatic reasoning performed by our system: static and dynamic occlusions are removed automatically, rendering information is collected, and semantic selection tools help guide user input. We also demonstrate that our user-assisted SfM method outperforms existing techniques on both real-world construction data and established multi-view datasets.
△ Less
Submitted 24 December, 2019;
originally announced December 2019.
-
Geometry-Informed Material Recognition
Authors:
Joseph DeGol,
Mani Golparvar-Fard,
Derek Hoiem
Abstract:
Our goal is to recognize material categories using images and geometry information. In many applications, such as construction management, coarse geometry information is available. We investigate how 3D geometry (surface normals, camera intrinsic and extrinsic parameters) can be used with 2D features (texture and color) to improve material classification. We introduce a new dataset, GeoMat, which…
▽ More
Our goal is to recognize material categories using images and geometry information. In many applications, such as construction management, coarse geometry information is available. We investigate how 3D geometry (surface normals, camera intrinsic and extrinsic parameters) can be used with 2D features (texture and color) to improve material classification. We introduce a new dataset, GeoMat, which is the first to provide both image and geometry data in the form of: (i) training and testing patches that were extracted at different scales and perspectives from real world examples of each material category, and (ii) a large scale construction site scene that includes 160 images and over 800,000 hand labeled 3D points. Our results show that using 2D and 3D features both jointly and independently to model materials improves classification accuracy across multiple scales and viewing directions for both material patches and images of a large scale construction site scene.
△ Less
Submitted 18 July, 2016;
originally announced July 2016.
-
Toward a Science of Autonomy for Physical Systems: Construction
Authors:
Miroslaw Skibniewski,
Mani Golparvar-Fard
Abstract:
Today, ensuring and improving safety, productivity, quality, and sustainability in construction, operation, and maintenance of national civil infrastructure systems through advances in robotics and automation is a national imperative. By "national civil infrastructure" we refer to the 4.5M commercial buildings, 3.9M miles of public roads, 2M miles of oil and natural gas pipelines, 600K bridges, 19…
▽ More
Today, ensuring and improving safety, productivity, quality, and sustainability in construction, operation, and maintenance of national civil infrastructure systems through advances in robotics and automation is a national imperative. By "national civil infrastructure" we refer to the 4.5M commercial buildings, 3.9M miles of public roads, 2M miles of oil and natural gas pipelines, 600K bridges, 190K cell phone towers, 120K miles of major railroads, 100K miles of levees, 84K dams, 50K miles of electrical power lines, 25K miles of commercially navigable waterways, and 5K public-use airports in the United States, all of which are critical to our national economy and society. By "ensuring and improving safety, productivity, quality, and sustainability" we mean to guarantee performance across the entire life cycle of civil infrastructure, from construction to operation and from maintenance to disposal. By "advances in robotics and automation" we mean innovative research and education that will provide tools for systematic and timely execution of construction, and also collection, analysis, visualization, and operational use of big data in built environments for project monitoring and control purposes. In the following, we discuss the current state of construction and operation of the national civil infrastructure systems in detail and present several opportunities for improvements through research and education on robotics and automation.
△ Less
Submitted 12 April, 2016;
originally announced April 2016.
