Using Barriers to Reduce the Sensitivity to Edge Miscalculations of Casting-Based Object Projection Feature Estimation
This addresses a specific issue in computer vision for markerless tracking systems, but appears incremental as it builds on existing casting-based methods.
The paper tackles the problem of edge miscalculations in 3D motion tracking by proposing a barrier extension to casting-based techniques, which reduces sensitivity to such errors without specifying concrete numerical results.
3D motion tracking is a critical task in many computer vision applications. Unsupervised markerless 3D motion tracking systems determine the most relevant object in the screen and then track it by continuously estimating its projection features (center and area) from the edge image and a point inside the relevant object projection (namely, inner point), until the tracking fails. Existing reliable object projection feature estimation techniques are based on ray-casting or grid-filling from the inner point. These techniques assume the edge image to be accurate. However, in real case scenarios, edge miscalculations may arise from low contrast between the target object and its surroundings or motion blur caused by low frame rates or fast moving target objects. In this paper, we propose a barrier extension to casting-based techniques that mitigates the effect of edge miscalculations.