Cudasift tuning and SSC supporting multicameras (#1677)

* CudaSIFT: filter doubles

* removed fixed threshold

* SSC can be used with multicameras. Refactored CudaSIFT to support SSC. Add new parameter SIFT/MaxGaussianThreshold. DbViewer: show negative features with gray color (so that we can know which features are in the vocabulary)

* Added SIFT/MaxGaussianThreshold parameter

* Updated parameter description
This commit is contained in:
matlabbe
2026-03-28 18:47:46 -07:00
committed by GitHub
parent 94bd3601fc
commit 20409d2bf6
11 changed files with 232 additions and 180 deletions
+1
View File
@@ -2296,6 +2296,7 @@ std::vector<int> SSC(
const std::vector<cv::KeyPoint> & keypoints, int maxKeypoints, float tolerance, int cols, int rows, const std::vector<int> & indx)
{
bool useIndx = keypoints.size() == indx.size();
maxKeypoints = maxKeypoints - round(maxKeypoints * tolerance); // Just the make sure the solution will always be <= input maxKeypoints
// several temp expression variables to simplify solution equation
int exp1 = rows + cols + 2*maxKeypoints;