W|h(CSrSSKJr SSKrSSKJrJr \\"S/5S Sj55r"SS5r "S S \ 5r \S S j5r g)z?Contains everything that can help automate the cuts in MoviePy.) defaultdictN)convert_parameter_to_secondsuse_clip_fps_by_default start_timec  ^U4Sjn[R"UTRSU- 5SSnU"S5nUVs/sH"n[R"XS"U55SPM$ nnU[R"U5$s snf)aFind the period of a video based on frames correlation. Parameters ---------- clip : moviepy.Clip.Clip Clip for which the video period will be computed. fps : int, optional Number of frames per second used computing the period. Higher values will produce more accurate periods, but the execution time will be longer. start_time : float, optional First timeframe used to calculate the period of the clip. Examples -------- .. code:: python from moviepy import * from moviepy.video.tools.cuts import find_video_period clip = VideoFileClip("media/chaplin.mp4").subclipped(0, 1).loop(2) round(videotools.find_video_period(clip, fps=80), 6) 1 cB>TRU5R5$N) get_frameflatten)tclips R/home/james-whalen/.local/lib/python3.13/site-packages/moviepy/video/tools/cuts.pyframe find_video_period..frame)s~~a ((**Nr)rr)nparangedurationcorrcoefargmax)r fpsrrtimingsrefr corrss` rfind_video_periodr sv>+ii DMM1s7;AB?G (C7> ?w!R[[eAh ' -wE ? 299U# $$ @s)Bc6\rSrSrSrSrSrSrSrSr Sr g ) FramesMatch2a+Frames match inside a set of frames. Parameters ---------- start_time : float Starting time. end_time : float End time. min_distance : float Lower bound on the distance between the first and last frames max_distance : float Upper bound on the distance between the first and last frames cFXlX lX0lX@lX!- Ulgr )rend_time min_distance max_distance time_span)selfrr!r"r#s r__init__FramesMatch.__init__Es"$ ((!.rcdSURURURUR4-$)Nz(%.04f, %.04f, %.04f, %.04f)rr!r"r#r%s r__str__FramesMatch.__str__Ls4- OO MM       1   rc"UR5$r )r+r*s r__repr__FramesMatch.__repr__Ts||~rcp[URURURUR45$r )iterrr!r"r#r*s r__iter__FramesMatch.__iter__Ws/ __dmmT->->@Q@Q R  rcURUR:H=(aY URUR:H=(a9 URUR:H=(a URUR:H$r r))r%others r__eq__FramesMatch.__eq__\sb    / 8$--/ 8""d&7&77 8""d&7&77  r)r!r#r"rr$N) __name__ __module__ __qualname____firstlineno____doc__r&r+r.r2r6__static_attributes__rrrr2s $/   rrcj\rSrSrSrSrS SjrSrSr\ S5r \ SS j5r SS jr S r S rg) FramesMatchesezFrames matches inside a set of frames. You can instantiate it passing a list of FramesMatch objects or using the class methods ``load`` and ``from_clip``. Parameters ---------- lst : list Iterable of FramesMatch objects. cB[RU[USS95 g)NcUR$r )r#)es r(FramesMatches.__init__..ssannrkey)listr&sorted)r%lsts rr&FramesMatches.__init__rs dF3,DEFrNcnUb[U5U-S- nUS:XaUS$[US[U55$)aOReturns a new instance of FramesMatches object or a FramesMatch from the current class instance given different conditions. By default returns the first FramesMatch that the current instance stores. Parameters ---------- n : int, optional Number of matches to retrieve from the current FramesMatches object. Only has effect when ``percent=None``. percent : float, optional Percent of the current match to retrieve. Returns ------- FramesMatch or FramesMatches : If the number of matches to retrieve is greater than 1 returns a FramesMatches object, otherwise a FramesMatch. Ndrr)lenr@int)r%npercents rbestFramesMatches.bestusA2  D G#c)Aq&tAwCmD3q6N&CCrc*[[X55$)aReturn a FramesMatches object obtained by filtering out the FramesMatch which do not satistify a condition. Parameters ---------- condition : func Function which takes a FrameMatch object as parameter and returns a bool. Examples -------- .. code:: python # Only keep the matches corresponding to (> 1 second) sequences. new_matches = matches.filter(lambda match: match.time_span > 1) )r@filter)r% conditions rrVFramesMatches.filters$VI455rc [R"U[R"UVs/sH"n[R"[U55PM$ sn5SSS9 gs snf)zSave a FramesMatches object to a file. Parameters ---------- filename : str Path to the file in which will be dumped the FramesMatches object data. z%.03f )fmt delimiterN)rsavetxtarrayrI)r%filenamerDs rsaveFramesMatches.savesD  HH6AbhhtAw'6 7  6s)Ac~[R"U5nUVs/sH n[U6PM nn[U5$s snf)zLoad a FramesMatches object from a file. Parameters ---------- filename : str Path to the file to use loading a FramesMatches object. Examples -------- >>> matching_frames = FramesMatches.load("somefile") )rloadtxtrr@)r_arrrDmfss rloadFramesMatches.loads:jj"(+,1{A,S!!-s:c |^^^URUR-S-mU4Sjm0mUU4Sjn/nURSUS9GHupxSUR5-n T"X5n [R "U 5n [ TR55H[n X|- U:aTRU 5 M[TU SU - 5TU SU -S.TU U'TU US U:TU US 'M] [TR55n XU S .TU'[U 5HupTU US (aMU"X|5nU=TU US 'TU US 'X:TU US 'XS -SHZnTUUTU Unn[US UUS -5US '[US UUS - US U- 5US 'US U:dMUSUS 'M\ M UTVs/sH5nUU:wdM TUUS (aMUUTUUS TUUS 4PM7 sn- nGM [UVs/sH n[U6PM sn5$s snfs snf)aFinds all the frames that look alike in a clip, for instance to make a looping GIF. Parameters ---------- clip : moviepy.video.VideoClip.VideoClip A MoviePy video clip. distance_threshold : float Distance above which a match is rejected. max_duration : float Maximal duration (in seconds) between two matching frames. fps : int, optional Frames per second (default will be ``clip.fps``). logger : str, optional Either ``"bar"`` for progress bar or ``None`` or any Proglog logger. Returns ------- FramesMatches All pairs of frames with ``end_time - start_time < max_duration`` and whose distance is under ``distance_threshold``. Examples -------- We find all matching frames in a given video and turn the best match with a duration of 1.5 seconds or more into a GIF: .. code:: python from moviepy import VideoFileClip from moviepy.video.tools.cuts import FramesMatches clip = VideoFileClip("foo.mp4").resize(width=200) matches = FramesMatches.from_clip( clip, distance_threshold=10, max_duration=3, # will take time ) best = matches.filter(lambda m: m.time_span > 1.5).best() clip.subclipped(best.start_time, best.end_time).write_gif("foo.gif") c.>X-R5T- $r )sum)F1F2N_pixelss r dot_product,FramesMatches.from_clip..dot_productsG==?X- -rc>T"TUSTUS5nTUSTUSpC[R"X4-SU-- 5$)Nr|F|sq)rsqrt)t1t2uvuvro frame_dicts rdistance)FramesMatches.from_clip..distancesSZ^G4jnW6MNBb>'*JrN7,Cq77151r6>* *rT) with_timeslogger?|F|)minmaxrrejected)rrrrrrN)wh iter_framesr rrtrIkeyspopabsrJ enumeraterrr@r)r distance_threshold max_durationrr~r{matching_framesr r flat_frame F_norm_sqF_normrvt_Fidistt3t3tt2t3rurDrnrorzs @@@r from_clipFramesMatches.from_clips`66DFF?Q& .  + ((D(HHAu}}.J#J;IWWY'F:??,-Fl*NN2& #:b>%#86#AB)"~e4v=)JrN1% #2q)%03EErN1%j1.*+C&0VTJqM"3b>!$Z0FJJ 2q!%(:b>!+"3EC!$SZU 1C!DCJ!$SZU 1CT%[SWEW!XCJ5z$66*.J'(, $ $B!GL%/^A%6z%BLQ 2q)%0*R.2CE2JK$  O[IfG!k1oGHH   Hs H4H4,H4H9c 0^ UcUn[S5nUHunm pxXVRT Xx/5 M [UR5SS9n /n Sn U GHuplXk:aM U V VVs/sHupoPM nnn nU V VVs/sHupnX:dM XU4PM nnn nUV VVs/sHupnX- U:dMXU4PM nnn nU(dMuU V VVs1sHupnXs:dM U iM nnn nUV s1sHoU- S::dM U iM nn UR U5(dM[ SU55m [ U 4SjU55um pxU R[UT Xx55 Xd-n GM [U 5$s snnn fs snnn fs snnn fs snnn fs sn f)a^Select the scenes at which a video clip can be reproduced as the smoothest possible way, mainly oriented for the creation of GIF images. Parameters ---------- match_threshold : float Maximum distance possible between frames. The smaller, the better-looping the GIFs are. min_time_span : float Minimum duration for a scene. Only matches with a duration longer than the value passed to this parameters will be extracted. nomatch_threshold : float, optional Minimum distance possible between frames. If is ``None``, then it is chosen equal to ``match_threshold``. time_distance : float, optional Minimum time offset possible between matches. Returns ------- FramesMatches : New instance of the class with the selected scenes. Examples -------- .. code:: python from pprint import pprint from moviepy import * from moviepy.video.tools.cuts import FramesMatches ch_clip = VideoFileClip("media/chaplin.mp4").subclipped(1, 4) mirror_and_clip = [ch_clip.with_effects([vfx.TimeMirror()]), ch_clip] clip = concatenate_videoclips(mirror_and_clip) result = FramesMatches.from_clip(clip, 10, 3).select_scenes( 1, 2, nomatch_threshold=0, ) print(result) # [(1.0000, 4.0000, 0.0000, 0.0000), # (1.1600, 3.8400, 0.0000, 0.0000), # (1.2800, 3.7200, 0.0000, 0.0000), # (1.4000, 3.6000, 0.0000, 0.0000)] c/$r r>r>rrrE-FramesMatches.select_scenes..qs"rc US$Nrr>)ks rrErus!rrGrg333333?c3*# UH upo1v M g7fr r>).0endr"r#s r .FramesMatches.select_scenes..sVCU?s,cCUsc3<># UHoST:XdM Uv M g7f)rNr>)rrDrs rrrs3-a1-s  ) rappendrJitems intersectionrnextrr@)r%match_threshold min_time_spannomatch_threshold time_distance dict_startsstartr"r# starts_endsresult min_startends_distancesrends great_matchesgreat_long_matches poor_matches short_matchess ` r select_scenesFramesMatches.select_scenes;sf  $ / !*- 6: 2E3   % %sL&G H7;[..0nE  %0 !E AOP=S CDP:H9G5S 12L19G :G"9F5S K=02L19F " &:H9G5S 39G  -1IDS5[S4HSDMI,,];;VCUVVC.23-3/ +C MM+eS,M N-II&1LV$$EQ " Js6'E7 ? E> E> !F 4F  F F - F=Fc ~UH7upE nSUSU-SU-4-nURXE5R"U40UD6 M9 g)anExtract the matching frames represented by the instance from a clip and write them as GIFs in a directory, one GIF for each matching frame. Parameters ---------- clip : video.VideoClip.VideoClip A video clip whose frames scenes you want to obtain as GIF images. gif_dir : str Directory in which the GIF images will be written. kwargs Passed as ``clip.write_gif`` optional arguments. Examples -------- .. code:: python import os from pprint import pprint from moviepy import * from moviepy.video.tools.cuts import FramesMatches ch_clip = VideoFileClip("media/chaplin.mp4").subclipped(1, 4) clip = concatenate_videoclips([ch_clip.time_mirror(), ch_clip]) result = FramesMatches.from_clip(clip, 10, 3).select_scenes( 1, 2, nomatch_threshold=0, ) os.mkdir("foo") result.write_gifs(clip, "foo") # MoviePy - Building file foo/00000100_00000400.gif with imageio. # MoviePy - Building file foo/00000115_00000384.gif with imageio. # MoviePy - Building file foo/00000128_00000372.gif with imageio. # MoviePy - Building file foo/00000140_00000360.gif with imageio. z%s/%08d_%08d.gifrNN) subclipped write_gif)r%r gifs_dirkwargsrr_names r write_gifsFramesMatches.write_gifssKP!% E1%3;c (JJD OOE ' 1 1$ A& A!%rr>)rN)Nbarr)r8r9r:r;r<r&rSrVr` staticmethodrfrrrr=r>rrr@r@es^ GD:6( """pIpIfUVd%L*Brr@c4Uc.URUSUS9Vs/sHoUR5PM nn[R"U[S9nUb UR nO[ U5SU- -n[[R"U55nUR5nS[R"[R"XrU-:55S-n S/[SU- U -5-U/-n [XSS5V V s/sHupX4PM n n n X4$s snfs sn n f)aDetects scenes of a clip based on luminosity changes. Note that for large clip this may take some time. Returns ------- tuple : cuts, luminosities cuts is a series of cuts [(0,t1), (t1,t2),...(...,tf)] luminosities are the luminosities computed for each frame of the clip. Parameters ---------- clip : video.VideoClip.VideoClip, optional A video clip. Can be None if a list of luminosities is provided instead. If provided, the luminosity of each frame of the clip will be computed. If the clip has no 'fps' attribute, you must provide it. luminosities : list, optional A list of luminosities, e.g. returned by detect_scenes in a previous run. luminosity_threshold : float, optional Determines a threshold above which the 'luminosity jumps' will be considered as scene changes. A scene change is defined as a change between 2 consecutive frames that is larger than (avg * thr) where avg is the average of the absolute changes between consecutive frames. logger : str, optional Either ``"bar"`` for progress bar or ``None`` or any Proglog logger. fps : int, optional Frames per second value. Must be provided if you provide no clip or a clip without fps attribute. Nuint32)rdtyper~)rrrr) rrkrr^floatrrOrdiffmeannonzerorIzip)r luminositiesluminosity_thresholdr~rfrluminosity_diffsavgluminosity_jumpsrrurvcutss r detect_scenesrsV!--#Xf-U UEEGU  88L6L mm,39-277<01    !C BHHRZZ 0#3M MN OPQ RRcD#)'7788C5@G#&w #< =#<RH#s D:D)Ng333333?)NN rN) r< collectionsrnumpyrmoviepy.decoratorsrrrrrIr@rr>rrrshE#T|n-#%.#%L0 0 ffBDfBR MQ;;r