Surveillance.layers.PuzzleScene Namespace Reference

Layered detector, track pointer, and perceiver classes for puzzle surveillance system. More...

Classes
class	CfgPuzzleScene
	Configuration instance for Puzzle Scene perceiver. Designed to work for processing subsets (detect, track, etc). More...
class	HandByDepth
	Detector based on depth only, no glove appearance model. More...
class	InstPuzzlePerceiver
	Class for collecting visual processing methods needed by the PuzzleScene perceiver. More...
class	InstPuzzleScene
	Class for collecting visual processing methods needed by the puzzle scene interpreter. More...
class	PuzzleActivities
	Simple puzzle activity monitor based on specified image regions. More...
class	PuzzleCalibrator
	Detection calibrator: Possible that not used due to static calibration methods in the Detector classes proper. More...
class	PuzzleDetectors
	Detector for layered puzzle scene: glove and puzzle pieces. More...
class	PuzzleMonitor
	Puzzle monitor that examines hand/glove state and puzzle state, or equivalent information as measured by a perceiver. More...
class	PuzzlePerceiver
	Perceiver based on glove and work scene/mat detection models. More...
class	StateDetectors
	Puzzle detector state information. More...
class	StatePuzzleActivity
	Perceiver puzzle scene activity state structure. More...
class	StatePuzzleScene
	Perceiver puzzle scene state aggegating detector and tracker, plus filter as fitting, information. More...
class	StatePuzzleTracks
	Basic puzzle trackpointer state information. More...
class	TrackPointers
	Track pointers for the glove/hand and the puzzle pieces. More...
class	ZoneCounter

Detailed Description

Layered detector, track pointer, and perceiver classes for puzzle surveillance system.

Given how python works regarding code import, it seems like the best would be to simply put all of the code into this one file. That will make it rather long, but with a consistent coding interface. If done properly, the individual classes will lean heavily on other code libraries and be relatively compact.

What should be contained in this file would be:

1. Layered detector from RGBD input.
2. Layered trackpointers based on layered detector output.
3. Layered perceiver that combine detector + trackpointers.
4. A calibration scheme for the entire process with saving to YAML and HDF5 files.

This single file replaces/supercedes the existing files in this directory (human_seg, robot_seg, tabletop_seg, puzzle_seg, base_bg, base_fg, base).

Author

Patricio A. Vela, pvela.nosp@m.@gat.nosp@m.ech.e.nosp@m.du

Date

2023/06/29