SRDCF_with_DenseNet
所属分类:人工智能/神经网络/深度学习
开发工具:matlab
文件大小:7351KB
下载次数:3
上传日期:2020-07-04 12:57:43
上 传 者:
zllxot
说明: 在SRDCF算法的基础上,将原来的fhog特征换成深度特征,提取特征使用的是DenseNet特征
(Based on SRDCF algorithm, the original fhog feature is replaced by deep feature, and densenet feature is used to extract feature)
文件列表:
SRDCF_with_DenseNet\average_feature_region.m (1067, 2020-06-30)
SRDCF_with_DenseNet\cconvmtx2.m (495, 2020-06-30)
SRDCF_with_DenseNet\compilemex_linux.m (407, 2020-06-30)
SRDCF_with_DenseNet\compilemex_win.m (323, 2020-06-30)
SRDCF_with_DenseNet\compile_mexResize.m (0, 2020-06-30)
SRDCF_with_DenseNet\demo.m (238, 2020-06-30)
SRDCF_with_DenseNet\dft2dfs_matrix.m (1199, 2020-06-30)
SRDCF_with_DenseNet\fhog.m (3036, 2020-06-30)
SRDCF_with_DenseNet\get_colorspace.m (1284, 2020-06-30)
SRDCF_with_DenseNet\get_deepfeature.m (941, 2020-06-30)
SRDCF_with_DenseNet\get_features.m (1887, 2020-06-30)
SRDCF_with_DenseNet\get_fhog.m (630, 2020-06-30)
SRDCF_with_DenseNet\get_pixels.m (729, 2020-06-30)
SRDCF_with_DenseNet\gradient2.m (1248, 2020-06-30)
SRDCF_with_DenseNet\gradientHist.m (3496, 2020-06-30)
SRDCF_with_DenseNet\gradientMag.m (2393, 2020-06-30)
SRDCF_with_DenseNet\gradientMex.cpp (19306, 2020-06-30)
SRDCF_with_DenseNet\gradientMex.mexa64 (23054, 2020-06-30)
SRDCF_with_DenseNet\gradientMex.mexw64 (30720, 2020-06-30)
SRDCF_with_DenseNet\init_net.m (129, 2020-06-30)
SRDCF_with_DenseNet\integralVecImage.m (1625, 2020-06-30)
SRDCF_with_DenseNet\libopencv_core.a (3590102, 2020-06-30)
SRDCF_with_DenseNet\libopencv_core.so (2009448, 2020-06-30)
SRDCF_with_DenseNet\libopencv_imgproc.a (3414242, 2020-06-30)
SRDCF_with_DenseNet\libopencv_imgproc.so (1830760, 2020-06-30)
SRDCF_with_DenseNet\load_video_info.m (690, 2020-06-30)
SRDCF_with_DenseNet\mexResize.cpp (1731, 2020-06-30)
SRDCF_with_DenseNet\mexResize.mexa64 (102795, 2020-06-30)
SRDCF_with_DenseNet\mexResize.mexw64 (55296, 2020-06-30)
SRDCF_with_DenseNet\mtimesx.c (55598, 2020-06-30)
SRDCF_with_DenseNet\mtimesx.m (12170, 2020-06-30)
SRDCF_with_DenseNet\mtimesx.mexa64 (263249, 2020-06-30)
SRDCF_with_DenseNet\mtimesx.mexw64 (321536, 2020-06-30)
SRDCF_with_DenseNet\mtimesx_build.m (16880, 2020-06-30)
SRDCF_with_DenseNet\mtimesx_RealTimesReal.c (295670, 2020-06-30)
SRDCF_with_DenseNet\mtimesx_sparse.m (3101, 2020-06-30)
SRDCF_with_DenseNet\MxArray.cpp (19883, 2020-06-30)
SRDCF_with_DenseNet\MxArray.hpp (36099, 2020-06-30)
SRDCF_with_DenseNet\opencv2\calib3d\calib3d.hpp (38377, 2020-06-30)
SRDCF_with_DenseNet\opencv2\contrib\contrib.hpp (39172, 2020-06-30)
... ...
This MATLAB code implements the SRDCF tracker [1].
Installation:
If the pre-compiled mexfiles do not work, the provided compilemex_[linux|windows] should compile them from the provided source code. Alternatively, you can try to modify them for your system.
Instructions;
* The "demo.m" script runs the tracker on the provided "Couple" sequence.
* The "run_SRDCF.m" function can be directly integrated to the Online Tracking Benchmark (OTB).
* The "run_SRDCF.m" contains the default parameters used to produce the results reported in the paper [1].
Contact:
Martin Danelljan
martin.danelljan@liu.se
Third party code used in the implementation of this tracker is:
* Piotrs image processing toolbox [2]
* mtimesx [3]
* opencv [4]
* lightspeed toolbox [5]
[1] Martin Danelljan, Gustav Hager, Fahad Shahbaz Khan and Michael Felsberg.
Learning Spatially Regularized Correlation Filters for Visual Tracking.
In Proceedings of the International Conference in Computer Vision (ICCV), 2015.
[2] Piotr Dollar.
"Piotr’s Image and Video Matlab Toolbox (PMT)."
http://vision.ucsd.edu/~pdollar/toolbox/doc/index.html.
[3] http://www.mathworks.com/matlabcentral/fileexchange/25977-mtimesx-fast-matrix-multiply-with-multi-dimensional-support
[4] http://opencv.org/
[5] http://research.microsoft.com/en-us/um/people/minka/software/lightspeed/
近期下载者:
相关文件:
收藏者: