video-tracker
Skill10 starsChanged 6 days agoMITView on GitHub
Turns an experiment video into reproducible physics data: tracks declared objects (CSRT with template fallback), computes the amplitude spectrum of the coordinate series, and runs physics-consistency probes. Experimental and unstable — every run must be checked against a known reference.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- JiamingShi-JsPhys/agent-skillstheir own version, first seen Nov 1, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- JiamingShi-JsPhys/agent-skills★ 4same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository22 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .github/copilot-instructions.mdCopilot instructions
- .cursor/rules/chelate-shell.mdcCursor rule
- 3d-modelerSkill
- ai-trace-removerSkill
- bibtex-normalizerSkill
- capstone-labmateSkill
- conflict-detectorSkill
- data-profilerSkill
- document-evolution-trackerSkill
- formula-validatorSkill
- fulltext-extractorSkill
- latex-exporterSkill
- literature-retrieverSkill
- longitudinal-analysisSkill
- ml-auditorSkill
- multi-dim-managerSkill
- quantitative-validatorSkill
- reference-auditorSkill
- sci-visualizerSkill
- security-guardSkill
- task-organizerSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 27
Run this exact form — do not read the source tree:
- Very specific: line 61
Never modify the raw video: All products go under process/.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
video-tracker
Extracts coordinates from a video, derives spectra, and probes physical consistency. It never modifies the raw video and requires every parameter to come from the config.
- Stage:
ingest· Sandbox: no · Dependencies:numpy,opencv(matplotliboptional for figures)
Quick Execution
Run this exact form — do not read the source tree:
python run.py --root <WS> --action probe -- --config cfg.jsonprobeverifies the config and the first frame before any tracking starts.- Then
track→fft→report;physicsadds the consistency probes. - Outputs land in
<WS>/process/vt_output/(plus the gate JSON).
When to use
- A recorded experiment must yield position/angle over time.
- A periodic signal must be recovered from tracked coordinates (FFT).
- Tracked data must be checked for physical plausibility.
When NOT to use
- Low-contrast, heavily occluded or deformable footage — expect failure.
- The measurement is available directly from the instrument.
Actions
--action | Does | Extra arguments |
probe | Validate config + first frame | --config <cfg> |
track | Track the declared objects | --config <cfg> |
fft | Amplitude spectrum of the tracked series | --config <cfg> |
physics | Physical-consistency probes | --config <cfg> --family <fam> |
report | Markdown report from the products | --config <cfg> |
demo | Run the bundled demo | --config <cfg> |
Hard constraints
- Never modify the raw video. All products go under
process/. - No interactive input. ROI, FPS, scale and parameters all come from the config.
- Deterministic. Same video + same config → identical bytes.
Outputs
<WS>/process/vt_output/<video>_tracking.csv—frame, time_s, x_px, y_px, confidence, reinit(+x_cm, y_cmwhen a scale is set).<video>_summary.json,<video>_fft.json,<video>_spectra.png,<video>_report.md,<video>_physics.json,batch_summary.json.<WS>/process/gates/video-tracker_<ts>.json— the gate result.
Failure handling
- A missing or unreadable video raises a clear error and exits
2. - An ROI outside the first frame fails at init, before tracking.
- A per-object tracker loss is recorded, not fatal; a failed re-acquire is recorded as
reinit = False.
Limitations
- Experimental / unstable — treat every run as preliminary.
- CSRT accuracy is target-specific; sub-pixel and optical-flow support are optional.
- CPU-bound: a 1080p 60 fps clip with two objects can take several minutes.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.