A new bottom-up visual saliency model, Graph-Based Visual Saliency (GBVS), is proposed. It consists of two steps: rst forming activation maps on certain feature channels, and then normalizing them in a way which highlights conspicuity and admits combination with other maps. The model is simple, and biologically plausible insofar as it is naturally parallelized. This model powerfully predicts human xations on 749 variations of 108 natural images, achieving 98% of the ROC area of a human-based control, whereas the classical algorithms of Itti & Koch ([2], [3], [4]) achieve only 84%.

gbvs
gbvs\algsrc
gbvs\algsrc\connectMatrix.m
gbvs\algsrc\distanceMatrix.m
gbvs\algsrc\formMapPyramid.m
gbvs\algsrc\getDims.m
gbvs\algsrc\graphsalapply.m
gbvs\algsrc\graphsalinit.m
gbvs\algsrc\indexmatrix.m
gbvs\algsrc\initGBVS.m
gbvs\algsrc\makeLocationMap.m
gbvs\algsrc\mexArrangeLinear.cc
gbvs\algsrc\mexArrangeLinear.mexa64
gbvs\algsrc\mexArrangeLinear.mexglx
gbvs\algsrc\mexArrangeLinear.mexmaci
gbvs\algsrc\mexArrangeLinear.mexmaci64
gbvs\algsrc\mexArrangeLinear.mexw32
gbvs\algsrc\mexArrangeLinear.mexw64
gbvs\algsrc\mexAssignWeights.cc
gbvs\algsrc\mexAssignWeights.mexa64
gbvs\algsrc\mexAssignWeights.mexglx
gbvs\algsrc\mexAssignWeights.mexmaci
gbvs\algsrc\mexAssignWeights.mexmaci64
gbvs\algsrc\mexAssignWeights.mexw32
gbvs\algsrc\mexAssignWeights.mexw64
gbvs\algsrc\mexColumnNormalize.cc
gbvs\algsrc\mexColumnNormalize.mexa64
gbvs\algsrc\mexColumnNormalize.mexglx
gbvs\algsrc\mexColumnNormalize.mexmaci
gbvs\algsrc\mexColumnNormalize.mexmaci64
gbvs\algsrc\mexColumnNormalize.mexw32
gbvs\algsrc\mexColumnNormalize.mexw64
gbvs\algsrc\mexSumOverScales.cc
gbvs\algsrc\mexSumOverScales.mexa64
gbvs\algsrc\mexSumOverScales.mexglx
gbvs\algsrc\mexSumOverScales.mexmaci
gbvs\algsrc\mexSumOverScales.mexmaci64
gbvs\algsrc\mexSumOverScales.mexw32
gbvs\algsrc\mexSumOverScales.mexw64
gbvs\algsrc\mexVectorToMap.cc
gbvs\algsrc\mexVectorToMap.mexa64
gbvs\algsrc\mexVectorToMap.mexglx
gbvs\algsrc\mexVectorToMap.mexmaci
gbvs\algsrc\mexVectorToMap.mexmaci64
gbvs\algsrc\mexVectorToMap.mexw32
gbvs\algsrc\mexVectorToMap.mexw64
gbvs\algsrc\namenodes.m
gbvs\algsrc\partitionindex.m
gbvs\algsrc\principalEigenvectorRaw.m
gbvs\algsrc\simpledistance.m
gbvs\algsrc\sparseness.m
gbvs\compile
gbvs\compile\cleanmex.m
gbvs\compile\gbvs_compile.m
gbvs\compile\gbvs_compile2.m
gbvs\demo
gbvs\demo\demonstration.m
gbvs\demo\flicker_motion_demo.m
gbvs\demo\simplest_demonstration.m
gbvs\gbvs.m
gbvs\gbvs_fast.m
gbvs\gbvs_install.m
gbvs\initcache
gbvs\initcache\17__24__m__2.mat
gbvs\initcache\18__24__m__2.mat
gbvs\initcache\21__32__m__2.mat
gbvs\initcache\23__24__m__2.mat
gbvs\initcache\23__32__m__2.mat
gbvs\initcache\24__18__m__2.mat
gbvs\initcache\24__23__m__2.mat
gbvs\initcache\24__24__m__2.mat
gbvs\initcache\24__32__m__2.mat
gbvs\initcache\27__32__m__2.mat
gbvs\initcache\27__40__m__2.mat
gbvs\initcache\28__32__m__2.mat
gbvs\initcache\29__30__m__2.mat
gbvs\initcache\30__28__m__2.mat
gbvs\initcache\30__29__m__2.mat
gbvs\initcache\30__40__m__2.mat
gbvs\initcache\32__24__m__2.mat
gbvs\initcache\32__28__m__2.mat
gbvs\initcache\32__31__m__2.mat
gbvs\initcache\32__32__m__2.mat
gbvs\initcache\35__40__m__2.mat
gbvs\initcache\40__30__m__2.mat
gbvs\initcache\40__38__m__2.mat
gbvs\initcache\40__40__m__2.mat
gbvs\ittikochmap.m
gbvs\makeGBVSParams.m
gbvs\readme.txt
gbvs\saltoolbox
gbvs\saltoolbox\attenuateBordersGBVS.m
gbvs\saltoolbox\makeGaborFilterGBVS.m
gbvs\saltoolbox\maxNormalizeStdGBVS.m
gbvs\saltoolbox\mexLocalMaximaGBVS.cc
gbvs\saltoolbox\mexLocalMaximaGBVS.mexa64
gbvs\saltoolbox\mexLocalMaximaGBVS.mexglx
gbvs\saltoolbox\mexLocalMaximaGBVS.mexmaci
gbvs\saltoolbox\mexLocalMaximaGBVS.mexmaci64
gbvs\saltoolbox\mexLocalMaximaGBVS.mexw32

• We are an exchange download platform that only provides communication channels. The downloaded content comes from the internet. Except for download issues, please Google on your own.
• The downloaded content is provided for members to upload. If it unintentionally infringes on your copyright, please contact us.
• Please use Winrar for decompression tools
• If download fail, Try it againg or Feedback to us.
• If downloaded content did not match the introduction, Feedback to us，Confirm and will be refund.
• Before downloading, you can inquire through the uploaded person information

Nothing．