#include "Channels/ChannelOpts.H"#include "Channels/ColorComputeTypes.H"#include "Channels/IntegerDecodeType.H"#include "Channels/MapCombineType.H"#include "Channels/OrientComputeTypes.H"#include "Component/ModelOptionDef.H"#include "Component/OptionManager.H"#include "Image/LevelSpec.H"#include "Image/fancynorm.H"
Go to the source code of this file.
Functions | |
| void | REQUEST_OPTIONALIAS_CHANNEL (OptionManager &m) |
| request all of the above-named OPT_ALIAS options | |
Variables | |
| const ModelOptionCateg | MOC_CHANNEL |
| Command-line options for ChannelBase and derivatives. | |
| const ModelOptionDef | OPT_UseOlderVersion |
| const ModelOptionDef | OPT_ChannelOutputRangeMax |
| const ModelOptionDef | OPT_ChannelOutputRangeMin |
| const ModelOptionDef | OPT_MapLevel |
| const ModelOptionDef | OPT_LevelSpec |
| const ModelOptionDef | OPT_MaxNormType |
| const ModelOptionDef | OPT_SingleChannelQueueLen |
| const ModelOptionDef | OPT_SingleChannelTimeDecay |
| const ModelOptionDef | OPT_NumOrientations |
| const ModelOptionDef | OPT_OriInteraction |
| const ModelOptionDef | OPT_OrientComputeType |
| const ModelOptionDef | OPT_GaborChannelIntensity |
| const ModelOptionDef | OPT_UseTrigTab |
| const ModelOptionDef | OPT_DirectionChannelTakeSqrt |
| const ModelOptionDef | OPT_DirectionChannelLowThresh |
| const ModelOptionDef | OPT_NumColorBands |
| const ModelOptionDef | OPT_NumSatBands |
| const ModelOptionDef | OPT_HueBandWidth |
| const ModelOptionDef | OPT_SatBandWidth |
| const ModelOptionDef | OPT_NumIntensityBands |
| const ModelOptionDef | OPT_IntensityBandWidth |
| const ModelOptionDef | OPT_NumDirections |
| const ModelOptionDef | OPT_NumSpatioTemporalDirections |
| const ModelOptionDef | OPT_NumSpatioTemporalSpeeds |
| const ModelOptionDef | OPT_NumOpticalFlowDirections |
| const ModelOptionDef | OPT_NumDirectionsFOE |
| const ModelOptionDef | OPT_FoeMSTChannelDelta |
| const ModelOptionDef | OPT_NumTheta |
| const ModelOptionDef | OPT_LJunctionChannelDelta |
| const ModelOptionDef | OPT_TJunctionChannelDelta |
| const ModelOptionDef | OPT_XJunctionChannelDelta |
| const ModelOptionDef | OPT_EndPointChannelDelta |
| const ModelOptionDef | OPT_NumLOrients |
| const ModelOptionDef | OPT_NumTOrients |
| const ModelOptionDef | OPT_NumXOrients |
| const ModelOptionDef | OPT_NumEOrients |
| const ModelOptionDef | OPT_LFullImplementation |
| const ModelOptionDef | OPT_TFullImplementation |
| const ModelOptionDef | OPT_XFullImplementation |
| const ModelOptionDef | OPT_EFullImplementation |
| const ModelOptionDef | OPT_SingleChannelUseSplitCS |
| const ModelOptionDef | OPT_SingleChannelSaveRawMaps |
| const ModelOptionDef | OPT_SingleChannelComputeFullPyramidForGist |
| const ModelOptionDef | OPT_SingleChannelSaveFeatureMaps |
| const ModelOptionDef | OPT_SingleChannelSaveOutputMap |
| const ModelOptionDef | OPT_ComplexChannelSaveOutputMap |
| const ModelOptionDef | OPT_ComplexChannelMapCombineType |
| const ModelOptionDef | OPT_ColorComputeType |
| const ModelOptionDef | OPT_DummyChannelFactor |
| const ModelOptionDef | OPT_TcorrChannelFrameLag |
| const ModelOptionDef | OPT_ScorrChannelRadius |
| const ModelOptionDef | OPT_SubmapAlgoType |
| const ModelOptionDef | OPT_GetSingleChannelStats |
| const ModelOptionDef | OPT_SaveStatsPerChannel |
| const ModelOptionDef | OPT_SaveStatsPerChannelFreq |
| const ModelOptionDef | OPT_GetSingleChannelStatsFile |
| const ModelOptionDef | OPT_GetSingleChannelStatsTag |
| const ModelOptionDef | OPT_UseH2SV1 |
| const ModelOptionDef | OPT_CompColorDoubleOppWeight |
| const ModelOptionDef | OPT_CompColorSingleOppWeight |
| const ModelOptionCateg | MOC_INTCHANNEL |
| Command-line options for integer math in the channel classes. | |
| const ModelOptionDef | OPT_IntChannelScaleBits |
| const ModelOptionDef | OPT_IntChannelOutputRangeMax |
| const ModelOptionDef | OPT_IntChannelOutputRangeMin |
| const ModelOptionDef | OPT_IntMathLowPass5 |
| const ModelOptionDef | OPT_IntMathLowPass9 |
| const ModelOptionDef | OPT_IntInputDecode |
| const ModelOptionDef | OPT_ALIASsaveChannelOutputs |
| Channel-related command-line option aliases. | |
| const ModelOptionDef | OPT_ALIASsaveChannelInternals |
| const ModelOptionCateg | MOC_VCX |
| Command-line options for raw VisualCortex. | |
| const ModelOptionDef | OPT_RawVisualCortexOutputFactor |
| const ModelOptionDef | OPT_RawVisualCortexNoise |
| const ModelOptionDef | OPT_VCXsaveOutTo |
| const ModelOptionDef | OPT_VCXloadOutFrom |
| const ModelOptionDef | OPT_RawVisualCortexChans |
| const ModelOptionDef | OPT_IntegerRawVisualCortexChans |
| const ModelOptionDef | OPT_RawVisualCortexSaveOutput |
| const ModelOptionDef | OPT_VCXuseMax |
| const ModelOptionDef | OPT_VCXweightThresh |
Definition in file ChannelOpts.C.
| void REQUEST_OPTIONALIAS_CHANNEL | ( | OptionManager & | m | ) |
request all of the above-named OPT_ALIAS options
Definition at line 669 of file ChannelOpts.C.
References OptionManager::requestOptionAlias().
Referenced by REQUEST_OPTIONALIAS_NEURO().
| const ModelOptionCateg MOC_CHANNEL |
{
MOC_SORTPRI_3, "Channel-Related Options" }
Command-line options for ChannelBase and derivatives.
Channel-related options
Definition at line 52 of file ChannelOpts.C.
{
MOC_SORTPRI_3, "Channel-Related Integer Math Options" }
Command-line options for integer math in the channel classes.
Channel-related integer-math options
Definition at line 484 of file ChannelOpts.C.
| const ModelOptionCateg MOC_VCX |
{
MOC_SORTPRI_3, "VisualCortex-Related Options" }
Command-line options for raw VisualCortex.
VisualCortex-related options
Definition at line 546 of file ChannelOpts.C.
| const ModelOptionDef OPT_ALIASsaveChannelInternals |
{ MODOPT_ALIAS, "ALIASsaveChannelInternals", &MOC_CHANNEL, OPTEXP_SAVE,
"Save all available channel internal maps",
"save-channel-internals", '\0', "",
"--save-raw-maps --save-feature-maps" }
Definition at line 539 of file ChannelOpts.C.
{ MODOPT_ALIAS, "ALIASsaveChannelOutputs", &MOC_CHANNEL, OPTEXP_SAVE,
"Save all channel outputs",
"save-channel-outputs", '\0', "",
"--save-featurecomb-maps --save-conspic-maps" }
Channel-related command-line option aliases.
Definition at line 533 of file ChannelOpts.C.
| const ModelOptionDef OPT_ChannelOutputRangeMax |
{ MODOPT_ARG(float), "ChannelOutputRangeMax", &MOC_CHANNEL, OPTEXP_CORE,
"Max of the channel's output range",
"chanout-max", '\0', "<float>", "10.0" }
Definition at line 83 of file ChannelOpts.C.
| const ModelOptionDef OPT_ChannelOutputRangeMin |
{ MODOPT_ARG(float), "ChannelOutputRangeMin", &MOC_CHANNEL, OPTEXP_CORE,
"Min of the channel's output range",
"chanout-min", '\0', "<float>", "0.0" }
Definition at line 89 of file ChannelOpts.C.
| const ModelOptionDef OPT_ColorComputeType |
{ MODOPT_ARG(ColorComputeType), "ColorComputeType", &MOC_CHANNEL, OPTEXP_CORE,
"Type of computation used to compute RG and BY color opponencies",
"color-comp-type", '\0', "<Standard|Simple|StandardFull>",
"Standard" }
Definition at line 402 of file ChannelOpts.C.
| const ModelOptionDef OPT_CompColorDoubleOppWeight |
{ MODOPT_ARG(double), "CompColorDoubleOppWeight", &MOC_CHANNEL, OPTEXP_CORE,
"Weight to assign to double-opponent channels in CompositeColorChannel",
"compcolor-double-weight", '\0', "<double>", "1.0" }
Definition at line 473 of file ChannelOpts.C.
| const ModelOptionDef OPT_CompColorSingleOppWeight |
{ MODOPT_ARG(double), "CompColorSingleOppWeight", &MOC_CHANNEL, OPTEXP_CORE,
"Weight to assign to single-opponent channels in CompositeColorChannel",
"compcolor-single-weight", '\0', "<double>", "1.0" }
Definition at line 479 of file ChannelOpts.C.
| const ModelOptionDef OPT_ComplexChannelMapCombineType |
{ MODOPT_ARG(MapCombineType), "ComplexChannelMapCombineType", &MOC_CHANNEL, OPTEXP_CORE,
"Strategy used by ComplexChannel for combining output maps from "
"subchannels. Default strategy is summation; alternatives include "
"pixel-wise max.",
"map-combine-type", '\0', "<Sum|Max>", "Sum" }
Definition at line 394 of file ChannelOpts.C.
| const ModelOptionDef OPT_ComplexChannelSaveOutputMap |
{ MODOPT_FLAG, "ComplexChannelSaveOutputMap", &MOC_CHANNEL, OPTEXP_SAVE,
"Save conspicuity maps from all complex channels (\"CO\")",
"save-conspic-maps", '\0', "", "false" }
Definition at line 388 of file ChannelOpts.C.
| const ModelOptionDef OPT_DirectionChannelLowThresh |
{ MODOPT_ARG(float), "DirectionChannelLowThresh", &MOC_CHANNEL, OPTEXP_CORE,
"Low threshold to allpy to eliminate small motion responses",
"direction-lowthresh", '\0', "<float>", "3.0" }
Definition at line 189 of file ChannelOpts.C.
| const ModelOptionDef OPT_DirectionChannelTakeSqrt |
{ MODOPT_FLAG, "DirectionChannelTakeSqrt", &MOC_CHANNEL, OPTEXP_CORE,
"Take square root of our Reichardt output if true",
"direction-sqrt", '\0', "", "false" }
Definition at line 183 of file ChannelOpts.C.
| const ModelOptionDef OPT_DummyChannelFactor |
{ MODOPT_ARG(float), "DummyChannelFactor", &MOC_CHANNEL, OPTEXP_CORE,
"Factor to multiply the input by, or 0.0 for nothing",
"dummychannel-factor", '\0', "<float>", "0.0" }
Definition at line 409 of file ChannelOpts.C.
| const ModelOptionDef OPT_EFullImplementation |
{ MODOPT_FLAG, "EFullImplementation", &MOC_CHANNEL, OPTEXP_CORE,
"Use full implementation of the end-point channel",
"use-full-e", '\0', "", "false" }
Definition at line 351 of file ChannelOpts.C.
| const ModelOptionDef OPT_EndPointChannelDelta |
{ MODOPT_ARG(uint), "EndPointChannelDelta", &MOC_CHANNEL, OPTEXP_CORE,
"Distance from current pixel at which the presence or absence of "
"features should be checked for, when computing junction filter "
"outputs in EndPointChannel (--vc-chans=E).",
"end-point-delta", '\0', "<uint>", "6" }
Definition at line 301 of file ChannelOpts.C.
| const ModelOptionDef OPT_FoeMSTChannelDelta |
{ MODOPT_ARG(uint), "FoeMSTChannelDelta", &MOC_CHANNEL, OPTEXP_CORE,
"Distance from current pixel at which the presence or absence of "
"features should be checked for, when computing junction filter "
"outputs in foeMstChannel (--vc-chans=B).",
"foeMst-delta", '\0', "<uint>", "6" }
Definition at line 262 of file ChannelOpts.C.
| const ModelOptionDef OPT_GaborChannelIntensity |
{ MODOPT_ARG(double), "GaborChannelIntensity", &MOC_CHANNEL, OPTEXP_CORE,
"Intensity coefficient for Gabor channel",
"gabor-intens", '\0', "<double>", "10.0" }
Definition at line 170 of file ChannelOpts.C.
| const ModelOptionDef OPT_GetSingleChannelStats |
{ MODOPT_FLAG, "GetSingleChannelStats", &MOC_CHANNEL, OPTEXP_SAVE,
"Save (append) several types of statistics for each single channel to sc_stats.txt",
"save-channel-stats", '\0', "", "false" }
Definition at line 435 of file ChannelOpts.C.
| const ModelOptionDef OPT_GetSingleChannelStatsFile |
{ MODOPT_ARG(std::string), "GetSingleChannelStatsFile", &MOC_CHANNEL,
OPTEXP_SAVE,
"File name to use for single channel stats file",
"save-channel-stats-name", '\0', "<string>", "sc_stats.txt" }
Definition at line 453 of file ChannelOpts.C.
| const ModelOptionDef OPT_GetSingleChannelStatsTag |
{ MODOPT_ARG(std::string), "GetSingleChannelStatsTag", &MOC_CHANNEL,
OPTEXP_SAVE,
"Tag name to use for single channel stats file",
"save-channel-stats-tag", '\0', "<string>", "NULL" }
Definition at line 460 of file ChannelOpts.C.
| const ModelOptionDef OPT_HueBandWidth |
{ MODOPT_ARG(float), "HueBandWidth", &MOC_CHANNEL, OPTEXP_CORE,
"Width of each hue band",
"sigma-hueband", '\0', "<float>", "30" }
Definition at line 207 of file ChannelOpts.C.
| const ModelOptionDef OPT_IntChannelOutputRangeMax |
{ MODOPT_ARG(int), "IntChannelOutputRangeMax", &MOC_INTCHANNEL, OPTEXP_CORE,
"Max of the integer channel's output range",
"int-chanout-max", '\0', "<int>", "32768" }
Definition at line 494 of file ChannelOpts.C.
| const ModelOptionDef OPT_IntChannelOutputRangeMin |
{ MODOPT_ARG(int), "IntChannelOutputRangeMin", &MOC_INTCHANNEL, OPTEXP_CORE,
"Min of the integer channel's output range",
"int-chanout-min", '\0', "<int>", "0" }
Definition at line 500 of file ChannelOpts.C.
| const ModelOptionDef OPT_IntChannelScaleBits |
{ MODOPT_ARG(uint), "IntChannelScaleBits", &MOC_INTCHANNEL, OPTEXP_CORE,
"Number of working bits to use in integer channels",
"int-chan-scale-bits", '\0', "<uint>", "30" }
Definition at line 488 of file ChannelOpts.C.
| const ModelOptionDef OPT_IntegerRawVisualCortexChans |
{ MODOPT_ARG_STRING, "IntegerRawVisualCortexChans", &MOC_VCX, OPTEXP_CORE,
"The string specifies a list of channels to use in your IntegerVisualCortex as a series of "
"letters from:\n"
" C: double-opponent color center-surround\n"
" I: intensity center-surround\n"
" O: orientation contrast\n"
" F: flicker center-surround\n"
" M: motion energy center-surround\n"
" U: foe\n"
" B: FoeMST\n"
"with possible optional weights (given just after the channel letter, "
"with a ':' in between). Note that this is a subset of what is "
"supported by --vc-chans for the standard (float) VisualCortex.\n"
"EXAMPLE: 'IO:5.0M' will use intensity (weight 1.0), orientation "
"(weight 5.0) and motion (weight 1.0) channels",
"ivc-chans", '\0', "<CIOFM>", "CFIOM" }
Definition at line 625 of file ChannelOpts.C.
| const ModelOptionDef OPT_IntensityBandWidth |
{ MODOPT_ARG(float), "IntensityBandWidth", &MOC_CHANNEL, OPTEXP_CORE,
"Width of each intensity band",
"sigma-intband", '\0', "<float>", "22" }
Definition at line 225 of file ChannelOpts.C.
| const ModelOptionDef OPT_IntInputDecode |
{ MODOPT_ARG(IntegerDecodeType), "IntDecodeType",
&MOC_INTCHANNEL, OPTEXP_CORE,
"Which type of integer decoding to apply to input frames to "
"extract the luminance and chrominance components. The 'rgb' "
"decoding strategy is used by default, in which the input frame "
"is converted to 8-bit RGB, and from there the luminance and "
"chrominance are computed separately. The 'video' decoding "
"strategy may be more efficient if the input frames are natively "
"in some YUV format, since the luminance and chrominance are "
"computed in a single pass over the YUV data, with the Y component "
"used for the luminance and Y-normalized U/Y and V/Y components "
"used for chrominance; note that 'video' decoding strategy will "
"give somewhat different results from the 'rgb' decoding.",
"int-input-decode", '\0', "<rgb|video>", "rgb" }
Definition at line 517 of file ChannelOpts.C.
| const ModelOptionDef OPT_IntMathLowPass5 |
{ MODOPT_ARG(std::string), "IntMathLowPass5", &MOC_INTCHANNEL, OPTEXP_CORE,
"Which integer lowpass5 implementation to use",
"int-math-lowpass5", '\0', "<lp5std|lp5optim>", "lp5optim" }
Definition at line 506 of file ChannelOpts.C.
| const ModelOptionDef OPT_IntMathLowPass9 |
{ MODOPT_ARG(std::string), "IntMathLowPass9", &MOC_INTCHANNEL, OPTEXP_CORE,
"Which integer lowpass9 implementation to use",
"int-math-lowpass9", '\0', "<lp9std|lp9optim>", "lp9optim" }
Definition at line 512 of file ChannelOpts.C.
| const ModelOptionDef OPT_LevelSpec |
{ MODOPT_ARG(LevelSpec), "LevelSpec", &MOC_CHANNEL, OPTEXP_CORE,
"LevelSpec to use in channels. This controls the range of "
"spatial scales used in constructing center-surround maps, and "
"also controls the scale of the channel output maps. cmin and "
"cmax are the lowest (largest) and highest (largest) pyramid "
"levels to be used for the center scale in center-surround "
"operations. A level of 0 is the bottom pyramid level at the "
"original dimensions of the input image; each level above 0 "
"is reduced by a factor of 2 in the x and y dimensions, so e.g. "
"level 4 is 16-fold reduced in x and y. delmin and delmax "
"represent the range of differences at which the surround level "
"is offset from the center level. maplev is the scale at which "
"channel output should be generated. For example, the default "
"setting of 2,4,3,4,4 will use center scales 2-4 with deltas "
"of 3 and 4, for six center/surround pairs of 2/5, 2/6, 3/6, 3/7, "
"4/7 and 4/8, and the channel output will be at scale 4.",
"levelspec", '\0', "<cmin,cmax,delmin,delmax,maplev>", "2,4,3,4,4" }
Definition at line 101 of file ChannelOpts.C.
| const ModelOptionDef OPT_LFullImplementation |
{ MODOPT_FLAG, "LFullImplementation", &MOC_CHANNEL, OPTEXP_CORE,
"Use full implementation of the l-junction channel",
"use-full-l", '\0', "", "false" }
Definition at line 333 of file ChannelOpts.C.
| const ModelOptionDef OPT_LJunctionChannelDelta |
{ MODOPT_ARG(uint), "LJunctionChannelDelta", &MOC_CHANNEL, OPTEXP_CORE,
"Distance from current pixel at which the presence or absence of "
"features should be checked for, when computing junction filter "
"outputs in LJunctionChannel (--vc-chans=L).",
"l-junction-delta", '\0', "<uint>", "6" }
Definition at line 277 of file ChannelOpts.C.
| const ModelOptionDef OPT_MapLevel |
{ MODOPT_ARG(unsigned int), "MapLevel", &MOC_CHANNEL, OPTEXP_CORE,
"The level of the saliency map in Channels that don't need LevelSpec",
"maplevel", '\0', "<uint>", "4" }
Definition at line 95 of file ChannelOpts.C.
| const ModelOptionDef OPT_MaxNormType |
{ MODOPT_ARG(MaxNormType), "MaxNormType", &MOC_CHANNEL, OPTEXP_CORE,
"Type of MaxNormalization to use",
"maxnorm-type", '\0', "<None|Maxnorm|Fancy|FancyFast|FancyOne|"
"FancyLandmark|Landmark|FancyWeak|Ignore|Surprise>", "Fancy" }
Definition at line 121 of file ChannelOpts.C.
| const ModelOptionDef OPT_NumColorBands |
{ MODOPT_ARG(unsigned int), "NumColorBands", &MOC_CHANNEL, OPTEXP_CORE,
"Number of hue-band channels",
"num-colorband", '\0', "<int>", "3" }
Definition at line 195 of file ChannelOpts.C.
| const ModelOptionDef OPT_NumDirections |
{ MODOPT_ARG(unsigned int), "NumDirections", &MOC_CHANNEL, OPTEXP_CORE,
"Number of direction-selective motion channels",
"num-directions", '\0', "<int>", "4" }
Definition at line 231 of file ChannelOpts.C.
| const ModelOptionDef OPT_NumDirectionsFOE |
{ MODOPT_ARG(unsigned int), "NumDirections", &MOC_CHANNEL, OPTEXP_CORE,
"Number of direction-selective foe channels",
"num-foedirections", '\0', "<int>", "4" }
Definition at line 255 of file ChannelOpts.C.
| const ModelOptionDef OPT_NumEOrients |
{ MODOPT_ARG(unsigned int), "NumEOrients", &MOC_CHANNEL, OPTEXP_CORE,
"Number of oriented end-point channels",
"num-e-orients", '\0', "<int>", "4" }
Definition at line 327 of file ChannelOpts.C.
| const ModelOptionDef OPT_NumIntensityBands |
{ MODOPT_ARG(unsigned int), "NumIntensityBands", &MOC_CHANNEL, OPTEXP_CORE,
"Number of intensity-band channels",
"num-intensityband", '\0', "<int>", "3" }
Definition at line 219 of file ChannelOpts.C.
| const ModelOptionDef OPT_NumLOrients |
{ MODOPT_ARG(unsigned int), "NumLOrients", &MOC_CHANNEL, OPTEXP_CORE,
"Number of oriented l-junction channels",
"num-l-orients", '\0', "<int>", "4" }
Definition at line 309 of file ChannelOpts.C.
| const ModelOptionDef OPT_NumOpticalFlowDirections |
{ MODOPT_ARG(unsigned int), "NumOpticalFlowDirections", &MOC_CHANNEL, OPTEXP_CORE,
"Number of directions for optical flow motion channels",
"num-opticalflowdirections", '\0', "<int>", "8" }
Definition at line 249 of file ChannelOpts.C.
| const ModelOptionDef OPT_NumOrientations |
{ MODOPT_ARG(unsigned int), "NumOrientations", &MOC_CHANNEL, OPTEXP_CORE,
"Number of oriented channels",
"num-orient", '\0', "<int>", "4" }
Definition at line 140 of file ChannelOpts.C.
| const ModelOptionDef OPT_NumSatBands |
{ MODOPT_ARG(unsigned int), "NumSatBands", &MOC_CHANNEL, OPTEXP_CORE,
"Number of saturation-band channels",
"num-satband", '\0', "<int>", "3" }
Definition at line 201 of file ChannelOpts.C.
| const ModelOptionDef OPT_NumSpatioTemporalDirections |
{ MODOPT_ARG(unsigned int), "NumSpatioTemporalDirections", &MOC_CHANNEL, OPTEXP_CORE,
"Number of directions for spatiotemporal motion channels",
"num-spatiotemporaldirections", '\0', "<int>", "8" }
Definition at line 237 of file ChannelOpts.C.
| const ModelOptionDef OPT_NumSpatioTemporalSpeeds |
{ MODOPT_ARG(unsigned int), "NumSpatioTemporalSpeeds", &MOC_CHANNEL, OPTEXP_CORE,
"Number of optimal speeds for spatiotemporal motion channels",
"num-spatiotemporalspeeds", '\0', "<int>", "3" }
Definition at line 243 of file ChannelOpts.C.
| const ModelOptionDef OPT_NumTheta |
{ MODOPT_ARG(unsigned int), "NumTheta", &MOC_CHANNEL, OPTEXP_CORE,
"Number of oriented disparity channels",
"num-theta", '\0', "<int>", "1" }
Definition at line 271 of file ChannelOpts.C.
| const ModelOptionDef OPT_NumTOrients |
{ MODOPT_ARG(unsigned int), "NumTOrients", &MOC_CHANNEL, OPTEXP_CORE,
"Number of oriented t-junction channels",
"num-t-orients", '\0', "<int>", "4" }
Definition at line 315 of file ChannelOpts.C.
| const ModelOptionDef OPT_NumXOrients |
{ MODOPT_ARG(unsigned int), "NumXOrients", &MOC_CHANNEL, OPTEXP_CORE,
"Number of oriented x-junction channels",
"num-x-orients", '\0', "<int>", "2" }
Definition at line 321 of file ChannelOpts.C.
| const ModelOptionDef OPT_OrientComputeType |
{ MODOPT_ARG(OrientComputeType), "OrientComputeType", &MOC_CHANNEL, OPTEXP_CORE,
"Type of computation used to compute orientations",
"oricomp-type", '\0', "<Steerable|Gabor|GaborEnergyNorm>",
"Steerable" }
Definition at line 163 of file ChannelOpts.C.
| const ModelOptionDef OPT_OriInteraction |
{ MODOPT_ARG_STRING, "OriInteraction", &MOC_CHANNEL, OPTEXP_CORE,
"This describes the way in which the different orientations "
"within the OrientationChannel interact with each other. "
"'None' is for no interaction (default); for 'SubtractMean', "
"the mean of all orientation pyramids is subtracted from each "
"orientation pyramid. You can determine your own interaction by "
"specifying a vector with interaction coefficients: 'c1,...,cn'"
" where n is the number of orientations. c1 is the coeffiecient "
"for an orientation itself, c2 for the one with the next higher "
"angle and so on. For instance, for n=4, 'None' is the same as "
"'1,0,0,0', and 'SubtractMean' is the same as "
"'0.75,-0.25,-0.25,-0.25'",
"ori-interaction", '\0', "<None|SubtractMean|c1,...,cn>",
"None" }
Definition at line 146 of file ChannelOpts.C.
| const ModelOptionDef OPT_RawVisualCortexNoise |
{ MODOPT_ARG(float), "RawVisualCortexNoise", &MOC_VCX, OPTEXP_CORE,
"Noise applied to outputs of VisualCortex after --vcx-outfac has been applied",
"vcx-noise", '\0', "<float>", "1.0e-12" }
Definition at line 557 of file ChannelOpts.C.
| const ModelOptionDef OPT_RawVisualCortexOutputFactor |
{ MODOPT_ARG(float), "RawVisualCortexOutputFactor", &MOC_VCX, OPTEXP_CORE,
"Factor applied to outputs of VisualCortex to scale them to Amps of "
"synaptic input currents to saliency map",
"vcx-outfac", '\0', "<float>", "1.0e-9" }
Definition at line 550 of file ChannelOpts.C.
| const ModelOptionDef OPT_RawVisualCortexSaveOutput |
{ MODOPT_FLAG, "RawVisualCortexSaveOutput", &MOC_VCX, OPTEXP_SAVE,
"Save output of visual cortex (=input to the saliency map) as a float "
"image in PFM format with absolute saliency values. This is good "
"for comparing several saliency maps obtained for different images. "
"See saliency/matlab/pfmreadmatlab.m for a program that reads PFM "
"images into Matlab.",
"save-vcx-output", '\0', "", "false" }
Definition at line 644 of file ChannelOpts.C.
| const ModelOptionDef OPT_SatBandWidth |
{ MODOPT_ARG(float), "SatBandWidth", &MOC_CHANNEL, OPTEXP_CORE,
"Width of each saturation band",
"sigma-satband", '\0', "<float>", "0.3" }
Definition at line 213 of file ChannelOpts.C.
| const ModelOptionDef OPT_SaveStatsPerChannel |
{ MODOPT_FLAG, "SaveStatsPerChannel", &MOC_CHANNEL, OPTEXP_SAVE,
"When saving channel stats, should we put the data for each channel into its own file?",
"save-stats-per-channel", '\0', "", "false" }
Definition at line 441 of file ChannelOpts.C.
| const ModelOptionDef OPT_SaveStatsPerChannelFreq |
{ MODOPT_FLAG, "SaveStatsPerChannelFreq", &MOC_CHANNEL, OPTEXP_SAVE,
"When saving channel stats, should we also save frequency data?",
"save-stats-per-channel-freq", '\0', "", "false" }
Definition at line 447 of file ChannelOpts.C.
| const ModelOptionDef OPT_ScorrChannelRadius |
{ MODOPT_ARG(int), "ScorrChannelRadius", &MOC_CHANNEL, OPTEXP_CORE,
"Radius (in saliency map pixels) to compute the spatial correlations",
"scorrchannel-radius", '\0', "<int>", "1" }
Definition at line 423 of file ChannelOpts.C.
| const ModelOptionDef OPT_SingleChannelComputeFullPyramidForGist |
{ MODOPT_FLAG, "SingleChannelComputeFullPyramidForGist", &MOC_CHANNEL, OPTEXP_SAVE,
"Save raw input maps (pyramid levels) from all single channels for gist"
"computation (\"SR\")",
"save-raw-maps-gist", '\0', "", "false" }
Definition at line 369 of file ChannelOpts.C.
| const ModelOptionDef OPT_SingleChannelQueueLen |
{ MODOPT_ARG(int), "SingleChannelQueueLen", &MOC_CHANNEL, OPTEXP_CORE,
"Queue length for channels",
"qlen", '\0', "<int>", "1" }
Definition at line 128 of file ChannelOpts.C.
| const ModelOptionDef OPT_SingleChannelSaveFeatureMaps |
{ MODOPT_FLAG, "SingleChannelSaveFeatureMaps", &MOC_CHANNEL, OPTEXP_SAVE,
"Save center-surround feature maps from all single channels (\"SF\")",
"save-feature-maps", '\0', "", "false" }
Definition at line 376 of file ChannelOpts.C.
| const ModelOptionDef OPT_SingleChannelSaveOutputMap |
{ MODOPT_FLAG, "SingleChannelSaveOutputMap", &MOC_CHANNEL, OPTEXP_SAVE,
"Save combined center-surround output maps from all single channels (\"SO\")",
"save-featurecomb-maps", '\0', "", "false" }
Definition at line 382 of file ChannelOpts.C.
| const ModelOptionDef OPT_SingleChannelSaveRawMaps |
{ MODOPT_FLAG, "SingleChannelSaveRawMaps", &MOC_CHANNEL, OPTEXP_SAVE,
"Save raw input maps (pyramid levels) from all single channels (\"SR\")",
"save-raw-maps", '\0', "", "false" }
Definition at line 363 of file ChannelOpts.C.
| const ModelOptionDef OPT_SingleChannelTimeDecay |
{ MODOPT_ARG(double), "SingleChannelTimeDecay", &MOC_CHANNEL, OPTEXP_CORE,
"Time decay for channel queues",
"qtime-decay", '\0', "<float>", "20.0" }
Definition at line 134 of file ChannelOpts.C.
| const ModelOptionDef OPT_SingleChannelUseSplitCS |
{ MODOPT_FLAG, "SingleChannelUseSplitCS", &MOC_CHANNEL, OPTEXP_CORE,
"Use split positive/negative center-surround computations",
"use-split-cs", '\0', "", "false" }
Definition at line 357 of file ChannelOpts.C.
| const ModelOptionDef OPT_SubmapAlgoType |
{ MODOPT_ARG_STRING, "SubmapAlgoType", &MOC_CHANNEL, OPTEXP_CORE,
"Name of the SubmapAlgorithm type to use in SingleChannel",
"submap-algo", '\0', "<Std|List>", "Std" }
Definition at line 429 of file ChannelOpts.C.
| const ModelOptionDef OPT_TcorrChannelFrameLag |
{ MODOPT_ARG(int), "TcorrChannelFrameLag", &MOC_CHANNEL, OPTEXP_CORE,
"Frame lag with which to compute the temporal correlations",
"tcorrchannel-framelag", '\0', "<int>", "1" }
Definition at line 416 of file ChannelOpts.C.
| const ModelOptionDef OPT_TFullImplementation |
{ MODOPT_FLAG, "TFullImplementation", &MOC_CHANNEL, OPTEXP_CORE,
"Use full implementation of the t-junction channel",
"use-full-t", '\0', "", "false" }
Definition at line 339 of file ChannelOpts.C.
| const ModelOptionDef OPT_TJunctionChannelDelta |
{ MODOPT_ARG(uint), "TJunctionChannelDelta", &MOC_CHANNEL, OPTEXP_CORE,
"Distance from current pixel at which the presence or absence of "
"features should be checked for, when computing junction filter "
"outputs in TJunctionChannel (--vc-chans=T).",
"t-junction-delta", '\0', "<uint>", "6" }
Definition at line 285 of file ChannelOpts.C.
| const ModelOptionDef OPT_UseH2SV1 |
{ MODOPT_FLAG, "UseH2SV1", &MOC_CHANNEL, OPTEXP_CORE,
"Use H2SV1 color type in H2SV rather than H2SV2",
"use-h2sv1", '\0', "", "false" }
Definition at line 467 of file ChannelOpts.C.
| const ModelOptionDef OPT_UseOlderVersion |
{ MODOPT_FLAG, "UseOlderVersion", &MOC_CHANNEL, OPTEXP_CORE,
"Use the older version where we normalize responses within all feature "
"types",
"use-older-version", '\0', "", "true" }
Definition at line 76 of file ChannelOpts.C.
| const ModelOptionDef OPT_UseTrigTab |
{ MODOPT_FLAG, "UseTrigTab", &MOC_CHANNEL, OPTEXP_CORE,
"Whether to accelerate trig operations by using table lookup, at the "
"expense of some loss of numerical precision",
"use-trig-tab", '\0', "", "false" }
Definition at line 176 of file ChannelOpts.C.
| const ModelOptionDef OPT_VCXloadOutFrom |
{ MODOPT_ARG_STRING, "VCXloadOutFrom", &MOC_VCX, OPTEXP_CORE,
"Load the raw VisualCortex output map from the designated MGZ file if "
"a filename is specified. Typically, the MGZ file should contain maps "
"which have been previously saved using the --vcx-save-out-to option. "
"CAUTION: When this option is in use, we will not compute much in "
"the VisualCortex, instead reading the precomputed results off the disk. "
"This means in particular that whichever channels you may have in your "
"VisualCortex will not affect its output.",
"vcx-load-out-from", '\0', "<filename.mgz>", "" }
Definition at line 571 of file ChannelOpts.C.
| const ModelOptionDef OPT_VCXsaveOutTo |
{ MODOPT_ARG_STRING, "VCXsaveOutTo", &MOC_VCX, OPTEXP_CORE,
"Save the raw VisualCortex output map to the designated MGZ file if "
"a filename is specified. The saved outputs can later be re-read using "
"the --vcx-load-out-from option.",
"vcx-save-out-to", '\0', "<filename.mgz>", "" }
Definition at line 563 of file ChannelOpts.C.
| const ModelOptionDef OPT_VCXuseMax |
{ MODOPT_FLAG, "VCXuseMax", &MOC_VCX, OPTEXP_SAVE,
"Use max across features instead of sum to yield the combined saliency map.",
"vcx-usemax", '\0', "", "false" }
Definition at line 654 of file ChannelOpts.C.
| const ModelOptionDef OPT_VCXweightThresh |
{ MODOPT_ARG(float), "VCXweightThresh", &MOC_VCX, OPTEXP_SAVE,
"Lower threshold on total channel weight when counting the number of non-zero "
"weight channels to decide whether or not to apply one last round of maxnorm "
"to the VisualCortex output, if the number of non-zero-weight channels is 2 or more.",
"vcx-weight-thresh", '\0', "<float>", "0.0" }
Definition at line 660 of file ChannelOpts.C.
| const ModelOptionDef OPT_XFullImplementation |
{ MODOPT_FLAG, "XFullImplementation", &MOC_CHANNEL, OPTEXP_CORE,
"Use full implementation of the x-junction channel",
"use-full-x", '\0', "", "false" }
Definition at line 345 of file ChannelOpts.C.
| const ModelOptionDef OPT_XJunctionChannelDelta |
{ MODOPT_ARG(uint), "XJunctionChannelDelta", &MOC_CHANNEL, OPTEXP_CORE,
"Distance from current pixel at which the presence or absence of "
"features should be checked for, when computing junction filter "
"outputs in XJunctionChannel (--vc-chans=X).",
"x-junction-delta", '\0', "<uint>", "6" }
Definition at line 293 of file ChannelOpts.C.
1.6.3