How computational neuroscience could help improving face recognition systems?

doi:10.1109/ICCKE.2014.6993453

Paper

How computational neuroscience could help improving face recognition systems?

Published Dec 22, 2014 · S. Karimimehr, M. Yazdchi

2014 4th International Conference on Computer and Knowledge Engineering (ICCKE)

UNKNOWN SJR score

3

Citations

0

Influential Citations

Full text Semantic Scholar

Abstract

Computational neuroscience is a growing discipline in science, which tries to understand the operations of human brain and inspire from it as a new computational paradigm. Face recognition is an important question both in pattern recognition and neuroscience. In the last few years, neuroscientists found many facts about object recognition in primate's brain. Here, we propose a cortex inspired face recognition system which uses some findings about the brain such as the operations of feature extractor cells in visual cortex and the function of attention in discarding distracting parts of the images. Now it is the time to merge the knowledge of learning systems with biological findings. The proposed method named Advanced Neurologically Inspires Face recognition (ANIF) system is compared with previous model NIF and some well-known face recognition algorithms within different datasets, which shows remarkable results.

Study Snapshot

Key takeawayComputational neuroscience can improve face recognition systems by incorporating visual cortex operations and attention functions, leading to remarkable results.

PopulationOlder adults (50-71 years)

Sample size24

MethodsObservational

OutcomesBody Mass Index projections

ResultsSocial networks mitigate obesity in older groups.

Sign up to use Study Snapshot

Consensus is limited without an account. Create an account or sign in to get more searches and use the Study Snapshot.

Create an account

Paper

How computational neuroscience could help improving face recognition systems?

Sign up to use Study Snapshot

References

Sparsity-Regularized HMAX for Visual Recognition

Sparse HMAX, incorporating sparse firing, outperforms the original HMAX model in object recognition, suggesting its potential for computer vision.

A novel face recognition system inspired by computational neuroscience

Our novel face recognition method, based on computational neuroscience, performs similarly to well-known methods in automatic face recognition with varying illumination and pose.

Invariant Visual Object and Face Recognition: Neural and Computational Bases, and a Model, VisNet

Invariant visual object and face recognition in the brain can be achieved through self-organizing learning based on temporal and spatial statistics of visual input, as demonstrated by the VisNet model.

Kernel Sparse Representation-Based Classifier

The kernel sparse representation-based classifier (KSRC) improves the performance of the sparse representation-based classifier (SRC) on face image data by mapping data into a high-dimensional kernel feature space and using kernel trick techniques to reduce dimensionality.

How Can Selection of Biologically Inspired Features Improve the Performance of a Robust Object Recognition Model?

Selecting biologically inspired features, such as object parts, using an evolutionary algorithm approach improves the performance of a robust object recognition model in diverse tasks.

Citations

Enhanced Brain Inspired Model for Face Categorization Using Mutual Information Maximization

The proposed method improves face categorization performance by over 5% compared to the original HMAX model, using mutual information maximization to select more relevant patches and remove redundant ones.

Virtual Reality in Computational Neuroscience

Virtual reality techniques enhance computational neuroscience by enhancing data visualization and analysis, identifying existing barriers and future improvements.