Difference between revisions of "Projects:Improve tracking of individual markers and marker patterns"
(Created page with "== Overview == Add a few sentences of overview of your project here. Remember that this is a Mediawiki and you can add all kinds of multimedia data (images, sounds, video .....") |
|||
| Line 1: | Line 1: | ||
== Overview == | == Overview == | ||
| − | + | * Input : Tracking data in .csv format (Tracker, Nexus), 3D backpack points, detected single marker points | |
| − | + | * Output : Combine the datasets, fill in gaps where only some but not all makers of the full pattern is detected, and filter random flips in output from Nexus | |
| + | |||
| + | Subprojects: | ||
| + | 2.1 Offline data processing; | ||
| + | 2.2 Online/quasi-real time solution working from the data stream | ||
== Contact == | == Contact == | ||
| − | + | * Mate Nagy, mnagy@orn.mpg.de | |
| + | * Hemal Naik, hnaik@orn.mpg.de | ||
| Line 34: | Line 39: | ||
== Suggested/tested approaches == | == Suggested/tested approaches == | ||
| − | + | * Possible approach: | |
| + | A. Backpack points can be transferred to coordinate system of VICON using output of tracker (6DOF pose transformation between VICON and Backpack coordinate system). Tracker has better optimization techniques than nexus therefore trajectories are more complete. The result is compared with 3D points given by Nexus and a combined dataset can be created. | ||
| + | B. Results from A is compared with .csv of unlabeled points (Tracker output of 3D all detections) to label all detected 3D points and remove ghost points. | ||
| + | C. The combined result from A and B can be used then to identify random flips in angles, positions and IDs. | ||
Revision as of 17:23, 20 May 2019
Contents
Overview
- Input : Tracking data in .csv format (Tracker, Nexus), 3D backpack points, detected single marker points
- Output : Combine the datasets, fill in gaps where only some but not all makers of the full pattern is detected, and filter random flips in output from Nexus
Subprojects: 2.1 Offline data processing; 2.2 Online/quasi-real time solution working from the data stream
Contact
- Mate Nagy, mnagy@orn.mpg.de
- Hemal Naik, hnaik@orn.mpg.de
Aims
List the aims of your project, or what you expect anyone taking up the project is supposed to hopefully achieve. The more specific, the better.
Estimated level of difficulty
If you have an estimate, classify level of difficulty according to the description of the CCU in the cluster proposal into
- Standard problems which just require applying existing methods (Hiwi level)
- Elaborate problems which require substantial adaptation or extension of existing methods (Master student level)
- Special problems which require research of entirely new methods and might lead to a paper or two (Ph.D. student level)
Maybe add a short clarification of what you believe are the main difficulties, and why you believe this is the right classification.
Provided data
Give a specific description of the datasets you provide or can provide which people need to use to solve your problem. If available and/or necessary, also suggest some means for reading the data format. If you can provide links to the data so people can download an take a look, all the better. Also list any known limitations, whether you can easily acquire/record new data, or any other useful information.
Note: Once the CCU server is up and running, datasets should be stored there for easy availability. See the howtos on storage for details.
Suggested/tested approaches
- Possible approach:
A. Backpack points can be transferred to coordinate system of VICON using output of tracker (6DOF pose transformation between VICON and Backpack coordinate system). Tracker has better optimization techniques than nexus therefore trajectories are more complete. The result is compared with 3D points given by Nexus and a combined dataset can be created. B. Results from A is compared with .csv of unlabeled points (Tracker output of 3D all detections) to label all detected 3D points and remove ghost points. C. The combined result from A and B can be used then to identify random flips in angles, positions and IDs.
