Skip to content
Snippets Groups Projects
Select Git revision
  • develop
  • feature/MatlabBackup
  • feature/MatlabWrapper
  • feature/deviceName
  • feature/linux
  • master default protected
  • timing
  • v2.02
  • v2.01
  • v2.0
  • Release_1.0
  • HIP_name_chnage
  • deletedOld
  • v1.0.2
  • PaperRelease
  • Release-0.0.6b
  • Release-0.0.5b
  • Release-0.0.4b
  • Release-0.0.3a
  • Release-0.0.2a
  • Release-0.0.1a
21 results
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.030Apr28278428Mar3Feb8Jan6Dec24Nov23Oct22131127Aug25159Jul4319Jun18176May52128Apr25241817161514128743231Mar28272524212024Feb21201586111Jan11Nov6131Oct29282725242221191611109328Sep2726242321121075230Aug6521ApplyPolyTransformation implementationAllstats for AddImageWithGetstats for AddImageWithAddImageWith implementationAllstats for AddConstGetstats for AddConstCreated a ridged registration function that returns a normalized covariance matrix back. Seems to be off by a factor of 0.5.Binary CommitGaussian filter no longer has quantization error between dimensions. Type casting happens now at the begging and then at the end instead of in the kernel.CudaAdd changed to CudaAddConst'AddConstant' implemented as CudaAddAdded a image tiling command. Mainly to test ROI correctness.Added a retriveImage function to CudaImageContainerFixed ROI setting and using in CudaImageContainerUpdated to CUDA 7.0Added explicit type casting to suppress warningsMerge remote-tracking branch 'origin/master' into ROICalculating image overlap for chunking more accurately. Mainly for the singleton dimension case.Fixed image reduction when using the median caseFixed bug in pointer arithmetic in linear un-mixerAllow the un-mixing kernel to go into negative values.Binary commitAdded a function to convert an index into a (X,Y,Z) coordinateSpelling fix Fixed logical operators in Vec Funcs by removing arrant constStarted to introduce a Ridged Registration algorithm that stays on the deviceStarting to play around with ROI selection within the Cuda Image ContainerAdded the ability to retrieve the amount of memory each GPU has in total and currently available.Removed unused variablesFixed Variance calculation to use individual cuda calls instead of helper functionsCuda sum does not send data to the device if the number of elements to add is less then a block sizeAdded mex Cleanup functionSuppress warnings changesFixed device set bug for NCVAdded device countPut MRF denoising back in the header fileFixed argument count bug in Normalized CovarianceFixed missing signatures in CWrappers for linear unmixingHeader format changeMoved the def file into the resource section of the solution