There was an error while loading. Please reload this page.
Say how to cite the package, now that it has a DOI Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Qx7SWwc4XGYgXNNKrcNAgP
Video Basics: a declared frame rate is not a measurement of time The page introduced the frame rate as one of a file's properties and left it there, which invites reading it as a fact. Everything the toolbox reports about when something happened is that number multiplied or divided by something, and a wrong one rescales the answer rather than raising. The numbers make the point better than the argument does. On a 29.97 fps clip whose movement has a true rhythm of 2.00 Hz, the file's own rate returns 120 beats a minute; told 25 the same code returns 100, told 20 it returns 80, told 15 it returns 60. Two scripts in one study held the rate as a constant of 20 and 15 against footage that was 25, and the error was invisible because it was consistent. Also the quieter version, which the corpus has met more than once: a header describes a file and does not guarantee its contents. A clip encoded at 25 fps from a source that changed five times a second declares 25, carries the right number of frames, lasts the right length, and repeats four fifths of them. Every header-based check agrees with itself. It matters most for anything measuring change between frames, which is most of this toolbox, since a repeated frame reads as stillness that was never there. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Qx7SWwc4XGYgXNNKrcNAgP
The QoM page: never average the raw acceleration magnitude A stationary accelerometer reads 1 g, so the time-average of the raw vector norm is 1 g plus a small movement term and varies mostly by calibration. It is not a tilt measure either, since rotating a vector does not change its norm. Measured on a standstill competition with 56 entrants ranked on that average: 11 scored below 1 g, which a calibrated sensor held still cannot do; the score correlated with band-limited quantity of motion at rho = 0.16, p = 0.25, while real movement spanned a factor of 7.8; and grouped by which of eight sensor units a competitor wore, the score separated the groups at H = 48.1, p = 3.4e-08 where movement did not, H = 6.8, p = 0.45. Rescoring on movement moves 40 of 56 more than ten places and empties the podium. Says which of the existing functions avoids it, and the two habits that generalise: check that a score's range is physically possible before asking whether it is interesting, and treat a measure that separates devices better than people as a measure of the devices. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Qx7SWwc4XGYgXNNKrcNAgP
Threshold cost: 0.2 percent on the full 345-clip corpus, superseding 0.4 from 83
Record the 2026-08-12 releases: OpenCV 5 pose, threshold cost, circ_corr deferral
Publish the 360 anglegram/AEM page and strip the history narration Page 14 (anglegrams and AEM overlays) loses its draft banner, gains real figures rendered with the toolbox, and joins the Home navigation in a new 360-and-joint-sessions group alongside the ambiscape page, which now links to it. The dated micromotion-migration notes on pages 11 and 12, the old-behaviour framing on pages 2 and 9, and the version gate on page 13 are gone; provenance sections state where functions come from without the evolution story. Page 7 states the actual 0.05 threshold default, page 10 scopes pick_peaks to the pulse and audio-feature detectors, and the prose drops first person and contractions throughout. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TZ1T9cWFjCD93GdVrYmM4m
Drop the PDF build leftovers and stop restating micromotion's API _all_md_combined_tmp.md is generated by mds_to_pdf.sh (cat $(ls *.md) > "$tmpfile") and deleted again at the end of a successful run, so it is a leftover from an interrupted one and regenerable. Home.md.backup is a superseded copy of Home.md and is in the history. The script now excludes the tmpfile from its own glob and removes it on any exit. _qom, _posture, _physio and _mocap moved to micromotion on 2026-07-29 and MGT only re-exports them. Restating their signatures here had already drifted: the wiki published accel_to_speed(highpass=0.3) where the code has 0.2, and missed group_qom's normalize= parameter added in 1.8.0. Those pages now name the functions and link to micromotion for the parameters. Also: cut the stale April 2026 release notes from Home in favour of the changelog, merged the pose method list into ch. 12 instead of repeating it in ch. 4, pointed ch. 3's output-file paragraph at ch. 9, fixed the wiki_pics image links (blob URLs do not render) and the dead deep link into musicalgestures/documentation/_filter.md. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Qx7SWwc4XGYgXNNKrcNAgP
The toolkit page published four signatures with the old band `band_limited_qom`, `group_qom`, `pose_qom` and `normalized_qom` were all listed with `lo=0.3`, and the first two with `hi=15.0`. Every default is `micromotion.BAND`, 0.2-5 Hz, and has been since 2026-07-31. The four are now taken from the installed package rather than retyped, and the band is stated once above them with a link to the micromotion reference, so the four cannot drift apart from each other again.
Unspace the dashes on Home.md The rule is the author's own, at https://www.arj.no/2022/05/13/em-dash-not-hyphen/. Only this page is in scope: the rest of the wiki is written by collaborators and was deliberately left untouched. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01GdJYswDvYoGDb8cRqbdp7B
Point at the musiscape wiki, which now exists Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01SZfFujAsH2Ebsd3Ezc7kHo
Plain prose: remove inline bold The convention across this documentation is that emphasis is carried by the sentence, not by `**`. Applied to the project's own documents earlier; the wikis are separate repositories and were missed. Code spans and fenced blocks are untouched, so `**kwargs` and `2**(-t/t_double)` still say what they said. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01SZfFujAsH2Ebsd3Ezc7kHo
Link the published camera-comparison dataset (10.5281/zenodo.21611765)
remap360 now validated on real GoPro/Theta footage
Add ch. 13 Working with ambiscape: shared clock, device sync, 360-camera workflow Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Consolidate sound–motion wiki pages; merge mocap/pose documentation Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0147io2kMk8M6jcNFNt1G96m
Add sound–motion analysis tool documentation (PRs #351-353) Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0147io2kMk8M6jcNFNt1G96m
Add motiondescriptors() to video-based processes Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Wiki: pose models download via urllib (drop wget instruction); add pose_center/pose_distance Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Document resample, pose_segments, pose_waterfall options, and audio–movement analysis Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Document waterfall axes=False option
Document 1.4.9 changes: overwrite default, pose_waterfall styles, beat_statistics, show keys Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Document pose_waterfall, MediaPipe default, marker trails, tempogram polish Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Use horizontal/vertical display keys
Add space-time visualisations and combined SSM to video method list
List new audio methods (mfcc, tempo, beat_statistics, chromagram, descriptors save_data)
List new video methods (heatmap, motiontempo, motionvectors, eulerian, sonomotiongram)
Trim wiki pages 2-6, 8-9 to summaries linking to docs site Content from these pages has been moved to the authoritative docs site at https://fourms.github.io/MGT-python/user-guide/. Each page is now a brief summary with a direct link to the corresponding docs site page. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Fix stale API and URL references in wiki - Home.md: replace wrong readthedocs URL with fourms.github.io/MGT-python/ - 2-Loading: thresh=0.15 → threshold=0.15 in motionvideo() example - 7-Filtering: thresh → threshold parameter name throughout Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Clarify wiki scope and add canonical docs references across chapters
Update wiki with April 2026 CI, GPU, docs, and testing changes