CSRT tracking
Track an initialized object with correlation filters weighted by channel and spatial reliability.
The previous position is a starting estimate, not the new target location.
Initialize a bounding box and appearance model.
The illustrative response map peaks near the new position. Ambiguous or occluded targets can produce weak or competing peaks.
Update position, scale and reliable appearance support.
Update position, scale and reliable appearance support.
Illustrative example. The stages explain the method; they are not a live OpenCV execution.
Try it on an image
Experiment at pixel level
Initialize a tracker on the input rectangle, then locate it in the second image. One update previews tracking; it cannot assess long-term drift.
The engine loads on your first run. Your images stay in this browser.
Scroll over either image to zoom at the pointer. Use the scrollbars to pan both views over the same relative area. Zoom is relative to the input; pixel coordinates belong to each image. Warps can change scene correspondence.
Pixel inspector RGBA · native values · matched scale · 9 × 9 output pixels
Select a pixel
Select a pixel
When to use it
Follow a selected object when a more detailed CPU tracker is acceptable.
How it works
- 01Initialize a bounding box and appearance model.
- 02Correlate the learned filters with the next search region.
- 03Update position, scale and reliable appearance support.
initialized box → local search → model update
What to tune
Initialize on a clear frame. Search padding, feature channels and scale settings trade work against robustness.
Where it breaks down
Trackers can drift or lose an occluded object. A successful update should still be checked against application-specific confidence and geometry.
TypeScript API
Open an entry for its exact overloads, parameter descriptions, result ownership and pinned upstream source.
All of these calls execute on the CPU. Native objects need explicit disposal. See matrices and ownership and build compatibility.
Related methods
Lucas-Kanade optical flow
Follow selected points by finding the small displacement that best aligns each local patch.
Farneback dense optical flow
Estimate a displacement vector at every pixel using local polynomial approximations.
DIS dense optical flow
Combine fast patch alignment with spatial propagation and dense refinement.
Dual TV-L1 optical flow
Estimate flow with a robust data term and a total-variation smoothness penalty.