PCA for gender estimation: Which eigenvectors contribute?

Date

2002-01-01

Author

Balci, K
Atalay, Mehmet Volkan

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

52
views

0
downloads

A pruning schema is applied to Multi-Layer Perceptron (MLP) gender classifer MLP uses eigenvector coeffcients of the face space created by Principal Component Analysis (PCA). We show that pruning improves the initial MLP performance by preserving the most effective input while eliminating most of the units and connections. Pruning is also used as a tool to monitor which eigenvectors contribute to gender estimation. In addition, by usage of FERET face database, we test the PCA approach on gender estimation task in a bigger setting than the previous experiments.

Subject Keywords

Face images

URI

https://hdl.handle.net/11511/55492

Conference Name

16th International Conference on Pattern Recognition (ICPR)

Collections

Department of Computer Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Face recognition using Eigenfaces and neural networks Akalın, Volkan; Severcan, Mete; Department of Electrical and Electronics Engineering (2003) A face authentication system based on principal component analysis and neural networks is developed in this thesis. The system consists of three stages; preprocessing, principal component analysis, and recognition. In preprocessing stage, normalization illumination, and head orientation were done. Principal component analysis is applied to find the aspects of face which are important for identification. Eigenvectors and eigenfaces are calculated from the initial face image set. New faces are projected onto ...
Accurate equivalent models of sandwich laminates with honeycomb core and composite face sheets via optimization involving modal behavior Cinar, Okan; Erdal Erdoğmuş, Merve; Kayran, Altan (2017-03-01) An approach is introduced for determining accurate two-dimensional equivalent laminated models of sandwich laminates with honeycomb core and composite facesheets by optimization involving modal behavior. The approach relies on minimizing the objective function which is defined as the sum of the square of the differences between the natural frequencies of the honeycomb sandwich laminate estimated by the finite element analysis of the 3D detailed model with the actual honeycomb core geometry and by the 2D equ...
Investigation of fracture behavior of steel/steel laminates Şimşir, Mehmet; Öztürk, Tayfur; Department of Metallurgical and Materials Engineering (2004) A study is carried out into fracture behavior of steel/steel laminates both experimentally and through finite element analysis (FEM). The laminates produced by hot pressing consisted of low carbon and medium carbon steels with two volume fractions; 0.41 and 0.81. Fracture toughness, JIC has been measured using partial unloading technique assuming a critical value of crack extension. The technique is initially applied to monolithic material and then to the laminates in crack divider orientation. Evaluation o...
Autoencoder-Based Error Correction Coding for One-Bit Quantization Balevi, Eren; Andrews, Jeffrey G. (2020-06-01) This paper proposes a novel deep learning-based error correction coding scheme for AWGN channels under the constraint of one-bit quantization in receivers. Specifically, it is first shown that the optimum error correction code that minimizes the probability of bit error can be obtained by perfectly training a special autoencoder, in which "perfectly" refers to converging the global minima. However, perfect training is not possible in most cases. To approach the performance of a perfectly trained autoencoder...
Non-equilibrium molecular dynamics of electromigration in aluminum and its alloys Şen, Fatih Gürçağ; Aydınol, Mehmet Kadri; Department of Metallurgical and Materials Engineering (2006) With constant miniaturization of integrated circuits, the current densities experienced in interconnects in electronic circuits has been multiplied. Aluminum, which is widely used as an interconnect material, has fast diffusion kinetics under low temperatures. Unfortunately, the combination of high current density and fast diffusion at low temperatures causes the circuit to fail by electromigration (EM), which is the mass transport of atoms due to the momentum transfer between conducting electrons and diffu...

Citation Formats

K. Balci and M. V. Atalay, “PCA for gender estimation: Which eigenvectors contribute?,” presented at the 16th International Conference on Pattern Recognition (ICPR), QUEBEC CITY, CANADA, 2002, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/55492.