A Performance Evaluation of Surface Normals-Based Descriptors for Recognition of Objects using CAD-Models

This paper describes a study and analysis of surface normal-base descriptors for 3D object recognition. Specifically, we evaluate the behaviour of descriptors in the recognition process using virtual models of objects created from CAD software. Later, we test them in real scenes using synthetic objects created with a 3D printer from the virtual models. In both cases, the same virtual models are used on the matching process to find similarity. The difference between both experiments is in the type of views used in the tests. Our analysis evaluates three subjects: the effectiveness of 3D descriptors depending on the viewpoint of camera, the geometry complexity of the model and the runtime used to do the recognition process and the success rate to recognize a view of object among the models saved in the database.

Are drawings dead? …and performance over aesthetics?

When the act of 'drawing' became what can only be called formalised, (whose growth can be said to have blossomed during the Renaissance), there developed a separation between the drawing and its procurement. Recently, David Ross Scheer, in his book ‘The Death of Drawing, Architecture in the Age of Simulation’ wrote: ‘…whereas architectural drawings exist to represent construction, architectural simulations exist to anticipate building performance.’ Meanwhile, Paolo Belardi, in his work ‘Why Architects Still Draw’ likens a drawing to an acorn, where he says: ‘It is the paradox of the acorn: a project emerges from a drawing – even from a sketch, rough and inchoate - just as an oak tree emerges from an acorn.’ He tells us that Giorgio Vasari would work late at night ‘seeking to solve the problems of perspective’ and he makes a passionate plea that this reflective process allows the concept to evolve, grow and/or develop. However, without belittling Belardi, the virtual model now needs this self-same treatment where it is nurtured, coaxed and encouraged to be the inchoate blueprint of the resultant oak tree. The model now too can embrace the creative process going through the first phase of preparation, where it focuses on the problem. The manipulation of the available material can then be incubated so that it is reasoned and generates feedback. This paper serves to align this shift in perception, methodologies and assess whether the 2D paper abstraction still has a purpose and role in today’s digital world!