Invariant 3D Shape Recognition using Predictive Modular Neural Networks
This work addresses invariant recognition for applications like gesture and action recognition, but appears incremental as it extends an existing model.
The authors tackled the problem of 3D invariant shape and texture recognition by generalizing the PREMONN model to handle functions of two variables and non-Euclidean spaces, enabling recognition without a reference point and under occlusion, with experimental results provided.
In this paper PREMONN (PREdictive MOdular Neural Networks) model/architecture is generalized to functions of two variables and to non-Euclidean spaces. It is presented in the context of 3D invariant shape recognition and texture recognition. PREMONN uses local relation, it is modular and exhibits incremental learning. The recognition process can start at any point on a shape or texture, so a reference point is not needed. Its local relation characteristic enables it to recognize shape and texture even in presence of occlusion. The analysis is mainly mathematical. However, we present some experimental results. The methods presented in this paper can be applied to many problems such as gesture recognition, action recognition, dynamic texture recognition etc.