29
edits
Changes
Jump to navigation
Jump to search
m
If you have * to generate training data and build an idea about how to approach initial incarnation of the problem, or have tried something already which did not work well, please provide details here. If available, link some papers or detector:** find valid segments of 3D trajectories ** use existing code which might provide to project 3D tracks into 2D images** find suitable bounding box and use image crop as a possible solution or algorithmtraining image** build database of these and retrain object detector, see [[CCU:Tutorials|Tutorials]].
→Suggested/tested approaches
== Suggested/tested approaches ==
