Skip to content

performer_count counts the typical framing where the widest is the hall - #384

Merged
alexarje merged 2 commits into
masterfrom
performer-count-typical
Sep 12, 2026
Merged

alexarje merged 2 commits into
masterfrom
performer-count-typical

Conversation

@alexarje

Copy link
Copy Markdown
Contributor

Summary

  • performer_count(..., stat="widest"|"typical"): median over framings for lecture halls and slide captures, maximum (as before) for concerts; low/high carry the other.
  • User guide: lecture halls (audience filter off, typical framing) and GPU decoding through ffmpeg_input_args.
  • CHANGELOG under Unreleased.

From the second test recording (a 3 h 24 min PhD defence): widest-framing counts were 6–8 from figures in the slides; typical-framing counts are 1, 1, 2, 2.

Test plan

  • tests/test_performers.py covers both statistics; tests/test_camera.py unchanged.
  • mypy clean on the module.

🤖 Generated with Claude Code

https://claude.ai/code/session_01PcjBSXbXjqdj1tmYHoLiKd

alexarje and others added 2 commits September 13, 2026 00:07
A concert's widest framing holds everyone on stage; a lecture hall's widest framing
holds the audience and the projected slides, whose figures are not the speakers.
`stat="typical"` takes the median over framings instead of the maximum, and the
guide says how to switch the audience filter off where there is no raised stage and
how to decode a long 1080p file on the GPU through the detectors' ffmpeg options.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PcjBSXbXjqdj1tmYHoLiKd
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PcjBSXbXjqdj1tmYHoLiKd
@alexarje
alexarje merged commit 0aa1149 into master Sep 12, 2026
11 of 12 checks passed
@alexarje
alexarje deleted the performer-count-typical branch September 12, 2026 22:48
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant