Exploiting the Circulant Structure of Tracking-by-Detection with Kernels

@inproceedings{Henriques2012ExploitingTC,
  title={Exploiting the Circulant Structure of Tracking-by-Detection with Kernels},
  author={Jo{\~a}o F. Henriques and Rui Caseiro and Pedro Martins and Jorge P. Batista},
  booktitle={ECCV},
  year={2012}
}
function k = dgk(x1, x2, sigma) % Eq. 16 c = fftshift(ifft2(fft2(x1) .* conj(fft2(x2)))); d = x1(:)'*x1(:) + x2(:)'*x2(:) 2*c; k = exp(-1 / sigma^2 * abs(d) / numel(x1)); end Training image x (current frame) and test image z (next frame) must be preprocessed with a cosine window. y has a Gaussian shape centered on the target. x, y and z are M-by-N matrices. All FFT operations are standard in MATLAB. Classifier Negative samples (far away) Frame N 
Highly Influential
This paper has highly influenced 198 other papers. REVIEW HIGHLY INFLUENTIAL CITATIONS
Highly Cited
This paper has 1,170 citations. REVIEW CITATIONS

From This Paper

Figures, tables, and topics from this paper.
634 Citations
23 References
Similar Papers

Citations

Publications citing this paper.
Showing 1-10 of 634 extracted citations

1,171 Citations

0100200300'12'14'16'18
Citations per Year
Semantic Scholar estimates that this publication has 1,171 citations based on the available data.

See our FAQ for additional information.

References

Publications referenced by this paper.
Showing 1-10 of 23 references

Similar Papers

Loading similar papers…