Robust real-time face detection

Viola, Paul A.; Jones, Michael

doi:10.1109/iccv.2001.937709

Cited by 3,197 publications

(4,899 citation statements)

References 5 publications

Supporting

Mentioning

4,782

Contrasting

Unclassified

105

Order By: Relevance

“…γ i to zeros. Under the assumption that both the primal and dual problems are feasible and the Slater's condition satisfies, strong duality holds between (5) and (8), which means that their solutions coincide such that one can obtain both solutions by solving either of them for many convex problems. If we know all the weak classifiers, i.e., the matrix H can be computed a priori, the original problem (5) can be easily solved (at least in theory) because it is an optimization problem with simple nonnegativeness constraints.…”

Section: Boostingmentioning

confidence: 99%

“…Because the primal variables correspond to the dual constraints, solving RMP is equivalent to solving a relaxed version of the dual problem. With a finite w, the set of constraints in the dual (8) are finite, and we can solve (8) that satisfies all the existing constraints. If we can prove that among all the constraints that we have not added to the dual problem, no single constraint is violated, then we can conclude that solving the restricted problem is equivalent to solving the original problem.…”

Section: Boostingmentioning

confidence: 99%

“…Quasic-Newton algorithms like L-BFGS-B [19] can be used to solve (5). In contrast, usually sophisticated primal-dual interior-point based algorithms are needed for solving the convex problem (8). All in all, (5) is easier to solve.…”

Section: Boostingmentioning

confidence: 99%

“…Much effort has been devoted to analyzing AdaBoost [2] and other boosting algorithms [3,4,5] due to their great success in both classification and regression when applied to a wide variety of computer vision and machine learning tasks (see [7,8] for example). Both theoretical and experimental results have shown that boosting algorithms have impressive generalization performance.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Fully corrective boosting with arbitrary loss and regularization

Shen

Hengel

2013

Neural Networks

View full text Add to dashboard Cite

We propose a general framework for analyzing and developing fully corrective boosting-based classifiers. The framework accepts any convex objective function, and allows any convex (for example, p -norm, p ≥ 1) regularization term. By placing the wide variety of existing fully corrective boosting-based classifiers on a common footing, and considering the primal and dual problems together, the framework allows direct comparison between apparently disparate methods. By solving the primal rather than the dual the framework is capable of generating efficient fully-corrective boosting algorithms without recourse to sophisticated convex optimization processes. We show that a range of additional boosting-based algorithms can be incorporated into the framework despite not being fully corrective. Finally, we provide an empirical analysis of the performance of a variety of the most significant boosting-based classifiers on a few machine learning benchmark datasets.

show abstract

Section: Boostingmentioning

confidence: 99%

Section: Boostingmentioning

confidence: 99%

Section: Boostingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Fully corrective boosting with arbitrary loss and regularization

Shen

Hengel

2013

Neural Networks

View full text Add to dashboard Cite

show abstract

“…The sum-table is a precomputed data structure that acts as a lookup table, dramatically reducing the number of multiplications or additions required to evaluate a given expression. More specifically, the sum-table is a discrete version of an integral image [21,22].…”

Section: Fft and Sum-table For Denominator Of Nccmentioning

confidence: 99%

Fast normalized cross correlation for motion tracking using basis functions

Hii

Hann

Chase

et al. 2006

Computer Methods and Programs in Biomedicine

View full text Add to dashboard Cite

Digital Image-based Elasto-tomography (DIET) is an emerging method for noninvasive breast cancer screening. Effective clinical application of the DIET system requires highly accurate motion tracking of the surface of an actuated breast with minimal computation. Normalized cross correlation (NCC) is the most robust correlation measure for determining similarity between points in two or more images providing an accurate foundation for motion tracking. However, even using fast fourier transform (FFT) methods, it is too computationally intense for rapidly managing several large images. A significantly faster method of calculating the NCC is presented that uses rectangular approximations in place of randomly placed landmark points or the natural marks on the breast. These approximations serve as an optimal set of basis functions that are automatically detected, dramatically reducing computational requirements. To prove the concept, the method is shown to be 37-150 times faster than the FFT-based NCC with the same accuracy for simulated data, a visco-elastic breast phantom experiment and human skin. Clinically, this approach enables thousands of randomly placed points to be rapidly and accurately tracked providing high resolution for the DIET system.

show abstract

Object Recognition

Andrade‐Cetto

Villamizar

1999

Wiley Encyclopedia of Electrical and Electronics Engineering

View full text Add to dashboard Cite

Robust real-time face detection

Cited by 3,197 publications

References 5 publications

Fully corrective boosting with arbitrary loss and regularization

Fully corrective boosting with arbitrary loss and regularization

Fast normalized cross correlation for motion tracking using basis functions

Object Recognition

Contact Info

Product

Resources

About