In the real world, a realistic setting for computer vision or multimedia recognition problems is that we have some classes containing lots of training data and many classes contain a small amount of training data. Therefore, how to use frequent classes to help learning rare classes for which it is harder to collect the training data is an open question. Learning with Shared Information is an emerging topic in machine learning, computer vision and multimedia analysis. There are different level of components that can be shared during concept modeling and machine learning stages, such as sharing generic object parts, sharing attributes, sharing transformations, sharing regularization parameters and sharing training examples, etc. Regarding the specific methods, multi-task learning, transfer learning and deep learning can be seen as using different strategies to share information. These learning with shared information methods are very effective in solving real-world large-scale problems. This special issue aims at gathering the recent advances in learning with shared information methods and their applications in computer vision and multimedia analysis. Both state-of-the-art works, as well as literature reviews, are welcome for submission. Papers addressing interesting real-world computer vision and multimedia applications are especially encouraged. Topics of interest include, but are not limited to:  • Multi-task learning or transfer learning for large-scale computer vision and multimedia analysis • Deep learning for large-scale computer vision and multimedia analysis • Multi-modal approach for large-scale computer vision and multimedia analysis • Different sharing strategies, e.g., sharing generic object parts, sharing attributes, sharing transformations, sharing regularization parameters and sharing training examples, • Real-world computer vision and multimedia applications based on learning with shared information, e.g., event detection, object recognition, object detection, action recognition, human head pose estimation, object tracking, location-based services, semantic indexing. • New datasets and metrics to evaluate the benefit of the proposed sharing ability for the specific computer vision or multimedia problem. • Survey papers regarding the topic of learning with shared information.  Authors who are unsure whether their planned submission is in scope may contact the guest editors prior to the submission deadline with an abstract, in order to receive feedback.

IEEE transactions on pattern analysis and machine intelligence

Recent developments in techniques for modeling, digitizing and visualizing 3D shapes has led to an explosion in the number of available 3D models on the Internet and in domain-specific databases. This has led to the development of 3D shape retrieval systems that, given a query object, retrieve similar 3D objects. For visualization, 3D shapes are often represented as a surface, in particular polygonal meshes, for example in VRML format. Often these models contain holes, intersecting polygons, are not manifold, and do not enclose a volume unambiguously. On the contrary, 3D volume models, such as solid models produced by CAD systems, or voxels models, enclose a volume properly. This paper surveys the literature on methods for content based 3D retrieval, taking into account the applicability to surface models as well as to volume models. The methods are evaluated with respect to several requirements of content based 3D shape retrieval, such as: (1) shape representation requirements, (2) properties of dissimilarity measures, (3) efficiency, (4) discrimination abilities, (5) ability to perform partial matching, (6) robustness, and (7) necessity of pose normalization. Finally, the advantages and limitations of the several approaches in content based 3D shape retrieval are discussed.

/pdf/a-survey-of-content-based-3d-shape-retrieval-methods-1p1e9hurpn.pdf

A survey of content based 3D shape retrieval methods

A Survey of Content Based 3D Shape Retrieval Methods (Figure 5).

In this paper, we investigate a data-driven synthesis approach to constructing 3D geometric surface models. We provide methods with which a user can search a large database of 3D meshes to find parts of interest, cut the desired parts out of the meshes with intelligent scissoring, and composite them together in different ways to form new objects. The main benefit of this approach is that it is both easy to learn and able to produce highly detailed geometric models -- the conceptual design for new models comes from the user, while the geometric details come from examples in the database. The focus of the paper is on the main research issues motivated by the proposed approach: (1) interactive segmentation of 3D surfaces, (2) shape-based search to find 3D models with parts matching a query, and (3) composition of parts to form new models. We provide new research contributions on all three topics and incorporate them into a prototype modeling system. Experience with our prototype system indicates that it allows untrained users to create interesting and detailed 3D models.

Modeling by example

Three-dimensional shape searching is a problem of current interest in several different fields. Most techniques have been developed for a particular domain and reduce a shape into a simpler shape representation. The techniques developed for a particular domain will also find applications in other domains. We classify and compare various 3D shape searching techniques based on their shape representations. A brief description of each technique is provided followed by a detailed survey of the state-of-the-art. The paper concludes by identifying gaps in current shape search techniques and identifies directions for future research.

https://engineering.purdue.edu/PRECISE/Publications/ThreeDimensionalShapeSearchingStateoftheartReviewan/review_shape.pdf

Three-dimensional shape searching: state-of-the-art review and future trends

This article describes the coarse shape filters that support the 3D, Internet-based search engine ShapeSifter, which aims to locate parts already in production that have a shape similar to a desired new part. The search engine analyzes the target model's shape characteristics and performs a similarity match on the database's contents. Our main challenge is identifying shape metrics that produce effective characterizations of 3D models for similarity comparison. We focus on using three novel convex hull-based indices to carry out a preliminary coarse filtering of candidates prior to more detailed analysis. We also describe the crucial role played by two databases of benchmark objects. Unlike other shape-matching systems, we chose ShapeSifter's architecture (multistep query), 3D representation (triangulated mesh), and implementation (Java3D) to allow deployment online once it matures.

Coarse filters for shape matching

Given a boundary representation, the problem of identifying sets of faces that bound holes and pockets in 2 1/2D objects is relatively trivial when they emanate from a single face, because the edges of the depression form an inner loop of convex edges. There are, however, many components in which a hole or pocket is present, and yet defines no inner loop on any face; in other words, it has a nonunique entrance face. The paper describes an algorithm, based on the manipulation of a face—edge graph, for identifying sets of faces in a 2 1/2D object that bound holes or pockets with unique or nonunique entrance faces. The algorithm also determines the perimeter of the projected area enclosed by each hole or pocket in a specific direction. This information is of direct use in the manufacture of the component.

Method for finding holes and pockets that connect multiple faces in 2 1/2D objects

This paper describes a system for the automatic recognition of assembly features and the generation of disassembly sequences. The paper starts by reviewing the nature and use of assembly features. One of the conclusions drawn from this survey is that the majority of assembly features involve sets of spatially adjacent faces. Two principle types of adjacency relationships are identified and an algorithm is presented for identifying assembly features which arise from "spatial" and "contact" face adjacency relationships (known as s-adjacency and c-adjacency respectively). The algorithm uses an octree representation of a B-rep model to support the geometric reasoning required to locate assembly features on disjoint bodies. A pointerless octree representation is generated by recursively sub-dividing the assembly model's bounding box into octants which are used to locate: 1. Those portions of faces which are c-adjacent (i.e. they effectively touch within the tolerance of the octree). 2. Those portions of faces which are s-adjacent to a nominated face. The resulting system can locate and partition spatially adjacent faces in a wide range of situations and at different resolutions. The assembly features located are recorded as attributes in the B-rep model and are then used to generate a disassembly sequence plan for the assembly. This sequence plan is represented by a transition state tree which incorporates knowledge of the availability of feasible gripping features. By way of illustration, the algorithm is applied to several trial components.

Automatic assembly feature recognition and disassembly sequence generation

Selecting optimal cutting tools that can answer to the performance criteria of manufacturing economics (quality, productivity, cost, etc) is an important step in planning the manufacture of components. Achieving this, however, is difficult because of the many constraints involved in the tool selection process. This paper describes a method for determining a theoretical optimal combination of cutting tools given a set of 3D volumes or 2D profiles. Optimal tools are selected by considering residual material that is inaccessible to oversized cutters and the relative clearance rates of cutters that can access these regions of the selected machining features. The current implementation described does not give exact results because several machining parameters have been ignored during the selection process, such as tool path length, plunge rates, etc. However, the experimental studies carried out to verify the theoretical results suggest that while these factors may influence the absolute values calculated, in ...

http://www1.gantep.edu.tr/~dereli/turkce/optitool.pdf

Optimizing tool selection

The subdivision of the removal, or delta-volume, of a machined component into meaningful volumes suitable for efficient machining is at present a complex and time-consuming task. An algorithm is presented for recognizing CNC machining volumes from 3D boundary representation solid models. The 2.5D volumes identified are associated with generic depressions and protrusions. The novelty of the algorithm lies in the fact that its success does not depend upon a pre-defined library of features and that it guarantees the accessibility of the volumes found.

D. E. R. Clark

Papers

Coarse filters for shape matching

Method for finding holes and pockets that connect multiple faces in 2 1/2D objects

Automatic assembly feature recognition and disassembly sequence generation

Optimizing tool selection

Delta-volume decomposition for multi-sided components