Skip with Aspect ratio only

To discuss the tuning of comskip and for posting your ini files
Post Reply
kihovuti
Posts: 7
Joined: Tue Dec 12, 2017 11:03 am

Skip with Aspect ratio only

Post by kihovuti » Tue Dec 12, 2017 11:14 am

Hi! I want to detect commercials based on aspect ratio only. It is detecting the blocks based on aspect ratio correctly but they are not being used to cut.

"Printing AR cblock list" shows 5 blocks. I want to keep block 0 and block 4.

Code: Select all

max_commercialbreak=1500
min_commercialbreak=25
max_commercial_size=180
min_commercial_size=4
min_show_segment_length=250
Detection methods log

Code: Select all

Detection Methods to be used:
	1) Scene Change
	2) Aspect Ratio


ComSkip throttles back from -0001 to -0001.
The time is now 1453 so it's full speed ahead!
End result log

Code: Select all

Input #0, mpegts, from 'd:\atv.ts':
  Duration: 00:08:42.31, start: 1.407833, bitrate: 248 kb/s
  Program 1 
    Metadata:
      service_name    : Service01
      service_provider: FFmpeg
    Stream #0:0[0x100]: Video: h264 (High) ([27][0][0][0] / 0x001B), yuv420p(progressive), 480x270, 25 fps, 25 tbr, 90k tbn, 50 tbc
    Stream #0:1[0x101]: Audio: aac (LC) ([15][0][0][0] / 0x000F), 48000 Hz, stereo, fltp, 36 kb/s
h.264 video can only be processed at full speed by the Donator version
h.264 video can only be processed at full speed by the Donator version
Frame Rate set to 25.000 f/s
Format changed to [480 : 270]
Frame: 1	Ratio: 2.06	MinY: 17 MaxY: 241 MinX: 1 MaxX: 480
Frame: 1 Channels:  0
Frame      2 (0.000s) - Black frame because large scene change of 96, uniform 52985

Initial audio pts =      0.000
Frame: 14 Channels:  2
Strange video pts step of 0.06450 instead of 0.04050 at frame 3046
Strange video pts step of 0.05233 instead of 0.04050 at frame 3067
Strange video pts step of 0.05629 instead of 0.04050 at frame 3089
Strange video pts step of 0.05147 instead of 0.04050 at frame 3101
Strange video pts step of 0.05413 instead of 0.04050 at frame 3117
Frame: 9195	Ratio: 1.79	MinY: 1	MaxY: 270	MinX: 1	MaxX: 480
Frame   9258 (371.120s) - Black frame because large scene change of 19, uniform 2894
Frame   9530 (382.000s) - Black frame because large scene change of 14, uniform 14404
Frame   9531 (382.040s) - Black frame because large scene change of 26, uniform 12198
Frame   9723 (389.720s) - Black frame because large scene change of 36, uniform 4388
Frame: 9978	Ratio: 2.14	MinY: 17	MaxY: 242	MinX: 1	MaxX: 480
Frame: 11040	Ratio: 1.85	MinY: 1	MaxY: 270	MinX: 1	MaxX: 480
Frame: 11175	Ratio: 2.14	MinY: 17	MaxY: 241	MinX: 1	MaxX: 480
Strange audio pts step of 9.05067 instead of 0.00000 at frame 12969
Jump in base apts from 509.96000 to 519.01010, delta=9.05010

Parsed 13032 video frames and 12828 audio frames at    25.75 fps

Maximum Volume found is 19745
Finished scanning file.  Starting to build Commercial List.
WARNING: Complex timeline or errors in the recording!!!!
Results may be wrong, .ref input will be misaligned. .txt editing will produce wrong results
Use .edl output if possible
Platau@[456] frames 8, volume 2, distance 18 seconds
Platau@[476] frames 8, volume 1, distance 0 seconds
Platau@[496] frames 7, volume 2, distance 0 seconds
Platau@[5055] frames 9, volume 2, distance 183 seconds
Platau@[5075] frames 7, volume 1, distance 0 seconds
Platau@[5095] frames 8, volume 0, distance 0 seconds
Platau@[12668] frames 21, volume 0, distance 302 seconds
Vol : #Frames
  0 : 7
Calculated silence level = 10
The last ar cblock wasn't closed.  Now closing.

Printing AR cblock list before cleaning
-----------------------------------------
Block: 0	Start:      1	End:   9194	AR_R: 2.06	Length: 0:06:08.60, [ 480x 270] minX=  1, minY= 17, maxX=480, maxY=241
Block: 1	Start:   9195	End:   9977	AR_R: 1.79	Length: 0:00:31.28, [ 480x 270] minX=  1, minY=  1, maxX=480, maxY=270
Block: 2	Start:   9978	End:  11039	AR_R: 2.14	Length: 0:00:42.43, [ 480x 270] minX=  1, minY= 17, maxX=480, maxY=242
Block: 3	Start:  11040	End:  11174	AR_R: 1.85	Length: 0:00:05.36, [ 480x 270] minX=  1, minY=  1, maxX=480, maxY=270
Block: 4	Start:  11175	End:  13032	AR_R: 2.14	Length: 0:01:14.24, [ 480x 270] minX=  1, minY= 17, maxX=480, maxY=241


After Sorting - 165
--------------
Aspect Ratio   2.14 found on  11240 frames totalling 	89.4%
Aspect Ratio   1.79 found on    849 frames totalling 	96.2%
Aspect Ratio   2.13 found on    282 frames totalling 	98.4%
Aspect Ratio   2.12 found on     82 frames totalling 	99.1%
Aspect Ratio   2.15 found on     43 frames totalling 	99.4%
Aspect Ratio   2.10 found on     21 frames totalling 	99.6%
Aspect Ratio   2.09 found on     14 frames totalling 	99.7%
Aspect Ratio   2.06 found on      8 frames totalling 	99.7%
Aspect Ratio   1.99 found on      7 frames totalling 	99.8%
Aspect Ratio   2.00 found on      5 frames totalling 	99.8%
Aspect Ratio   1.74 found on      3 frames totalling 	99.9%
Aspect Ratio   1.84 found on      3 frames totalling 	99.9%
Aspect Ratio   1.87 found on      2 frames totalling 	99.9%
Aspect Ratio   2.01 found on      2 frames totalling 	99.9%
Aspect Ratio   2.02 found on      2 frames totalling 	99.9%
Aspect Ratio   2.05 found on      2 frames totalling 	100.0%
Aspect Ratio   1.83 found on      1 frames totalling 	100.0%
Aspect Ratio   1.85 found on      1 frames totalling 	100.0%
Aspect Ratio   1.86 found on      1 frames totalling 	100.0%
Aspect Ratio   1.96 found on      1 frames totalling 	100.0%
Aspect Ratio   1.97 found on      1 frames totalling 	100.0%
Aspect Ratio   2.03 found on      1 frames totalling 	100.0%


After Sorting - 164
--------------
Aspect Ratio   2.06 found on   9194 frames totalling 	70.5%
Aspect Ratio   2.14 found on   2920 frames totalling 	93.0%
Aspect Ratio   1.79 found on    918 frames totalling 	100.0%

Printing AR cblock list
-----------------------------------------
Block: 0	Start:      1	End:   9194	AR_R: 2.06	Length: 0:06:08.60, [ 480x 270] minX=  1, minY= 17, maxX=480, maxY=241
Block: 1	Start:   9195	End:   9977	AR_R: 1.79	Length: 0:00:31.28, [ 480x 270] minX=  1, minY=  1, maxX=480, maxY=270
Block: 2	Start:   9978	End:  11039	AR_R: 2.14	Length: 0:00:42.43, [ 480x 270] minX=  1, minY= 17, maxX=480, maxY=242
Block: 3	Start:  11040	End:  11174	AR_R: 1.79	Length: 0:00:05.36, [ 480x 270] minX=  1, minY=  1, maxX=480, maxY=270
Block: 4	Start:  11175	End:  13032	AR_R: 2.14	Length: 0:01:14.24, [ 480x 270] minX=  1, minY= 17, maxX=480, maxY=241
Single frames with missing audio: 2
4 - Removing black frame 9723, from black frame list because volume 5000 is more than 500, brightness 23, uniform 4388
3 - Removing black frame 9531, from black frame list because volume 5000 is more than 500, brightness 176, uniform 12198
2 - Removing black frame 9530, from black frame list because volume 5000 is more than 500, brightness 199, uniform 14404
1 - Removing black frame 9258, from black frame list because volume 5000 is more than 500, brightness 23, uniform 2894
Distribution of Black Frame   cutting:   1 positive and   0 negative, ratio is 9.9900
Confidence of Black Frame   cutting:   0 out of   0 are strict, too low
Black Frame cutting too low
Distribution of Scene Change  cutting:   1 positive and   0 negative, ratio is 9.9900
Confidence of Scene Change  cutting:   0 out of   0 are strict, too low
Distribution of Uniform Frame cutting:   0 positive and   0 negative, ratio is 9.9900
Confidence of Uniform Frame cutting:   0 out of   0 are strict, too low
Black Frame List
---------------------------
Black Frame Count = 1
nr 	frame	pts	bright	uniform	volume		cause	dimcount  bright   type
  0	 13032	 522.080	     0	     0	     0	           b 	     0	     0	P
Creating cblock 0 From 1 (1) to 13032 (13032) because of                      b  with 0 head and 0 tail
Block 000	schange_rate - 0.12	 average - 0.12

Fuzzy scoring of the blocks
---------------------------
Block 0 has twice excess length.
Block 0 score:	Before - 1.00	After - 0.00
Threshold used - 1.0500	After rounding - 1.0500


	---------------------
	Initial Commercial List
	---------------------
No change

Logo fraction:              0.0000      Logo disabled
Maximum volume found:        19745
Average volume:               4536
Sound threshold:               500
Silence threshold:             100
Minimum volume found:           50
Average frames with silence:    90
Black threshold:                25
Minimum brightness found:       16
Minimum bright pixels found:   400
Minimum dim level found:       315
Average brightness:            731
Uniformity level:            11200
Average non uniformity:        253
Maximum gap between logo's:     -1
Suggested logo_threshold:   0.0000
Suggested shrink_logo:	    0.00
Max commercial size found:      -1
Dominant aspect ratio:      2.0600
Score threshold:            1.0500
Framerate:                  25.000
Average framerate:          24.960
Total commercial length:    0:00:00.00
Cut codes:
  F: scene	 c: change
  A: aspect	 t: cutscene
  E: exceeds	 l: logo
  L: logo	 v: volume
  B: bright	 s: scene_change
  C: combined	 a: aspect_ratio
  N: nonstrict	 u: uniform_frame
  S: strict	 b: black_frame
  		 r: resolution
----------------------------------------------------
Block list after weighing
----------------------------------------------------
  #     sbf  bs  be     fs     fe        ts        te       len     sc   scr cmb   ar                   cut    bri logo   vol sil   corr stdev   cc
  0:+     0   0   0      1  13032     0.00s   522.08s   522.08s   0.00  0.12   0 2.06            E         b   7310 0.00 45360 900  0.000    73           

13032 Frames Processed
################################################################
Time at end of run:
Tue Dec 12 15:07:07 2017
################################################################

---------------------

After above I changed

Code: Select all

min_show_segment_length=70
but still got the same result.

Then I turned on Logo detection (Logo Detection + Aspect Ratio). Now aspect ratio is detected correctly and consistently as you can see, but it is still not making any cut points when AR changes.

Code: Select all

Printing AR cblock list
-----------------------------------------
Block: 0	Start:      1	End:      1	AR_R: 2.14	Length: 0:00:00.00, [ 480x 270] minX=  1, minY= 29, maxX=480, maxY=241
Block: 1	Start:      2	End:   9194	AR_R: 2.26	Length: 0:06:08.56, [ 480x 270] minX=  1, minY= 29, maxX=480, maxY=241
Block: 2	Start:   9195	End:   9977	AR_R: 1.79	Length: 0:00:31.28, [ 480x 270] minX=  1, minY=  1, maxX=480, maxY=270
Block: 3	Start:   9978	End:  11040	AR_R: 2.26	Length: 0:00:42.48, [ 480x 270] minX=  1, minY= 30, maxX=480, maxY=242
Block: 4	Start:  11041	End:  11174	AR_R: 1.79	Length: 0:00:05.31, [ 480x 270] minX=  1, minY=  1, maxX=480, maxY=270
Block: 5	Start:  11175	End:  13032	AR_R: 2.26	Length: 0:01:14.24, [ 480x 270] minX=  1, minY= 30, maxX=480, maxY=241
Single frames with missing audio: 2
Distribution of Black Frame   cutting:   1 positive and   0 negative, ratio is 9.9900
Confidence of Black Frame   cutting:   0 out of   0 are strict, too low
Black Frame cutting too low
Distribution of Scene Change  cutting:   0 positive and   0 negative, ratio is 9.9900
Confidence of Scene Change  cutting:   0 out of   0 are strict, too low
Distribution of Uniform Frame cutting:   0 positive and   0 negative, ratio is 9.9900
Confidence of Uniform Frame cutting:   0 out of   0 are strict, too low
Black Frame List
---------------------------
Black Frame Count = 1
nr 	frame	pts	bright	uniform	volume		cause	dimcount  bright   type
  0	 13032	 522.080	     0	     0	     0	           b 	     0	     0	P
Creating cblock 0 From 1 (1) to 13032 (13032) because of                      b  with 0 head and 0 tail

Fuzzy scoring of the blocks
---------------------------
Block 0 has twice excess length.
Block 0 score:	Before - 1.00	After - 0.00
Threshold used - 1.0500	After rounding - 1.0500


	---------------------
	Initial Commercial List
	---------------------
No change

Logo fraction:              0.9982      Logo disabled
Maximum volume found:        19745
Average volume:               4536
Sound threshold:               500
Silence threshold:             100
Minimum volume found:           50
Average frames with silence:    90
Black threshold:                25
Minimum brightness found:       16
Minimum bright pixels found:     0
Minimum dim level found:         0
Average brightness:            745
Uniformity level:            11000
Average non uniformity:        258
Maximum gap between logo's:     -1
Suggested logo_threshold:   0.8250
Suggested shrink_logo:	    0.00
Max commercial size found:      -1
Dominant aspect ratio:      2.2600
Score threshold:            1.0500
Framerate:                  25.000
Average framerate:          24.960
Total commercial length:    0:00:00.00
Cut codes:
  F: scene	 c: change
  A: aspect	 t: cutscene
  E: exceeds	 l: logo
  L: logo	 v: volume
  B: bright	 s: scene_change
  C: combined	 a: aspect_ratio
  N: nonstrict	 u: uniform_frame
  S: strict	 b: black_frame
  		 r: resolution
----------------------------------------------------
Block list after weighing
----------------------------------------------------
  #     sbf  bs  be     fs     fe        ts        te       len     sc   scr cmb   ar                   cut    bri logo   vol sil   corr stdev   cc
  0:+     0   0   0      1  13032     0.00s   522.08s   522.08s   0.00  0.12   0 2.26            E         b   7450 0.00 45360 900  0.000    74           

Last edited by kihovuti on Tue Dec 12, 2017 12:22 pm, edited 1 time in total.

erik
Site Admin
Posts: 3310
Joined: Sun Aug 21, 2005 3:49 pm

Re: Skip with Aspect ratio only

Post by erik » Wed Dec 13, 2017 11:45 am

Did you enable cut_on_ar_change?

kihovuti
Posts: 7
Joined: Tue Dec 12, 2017 11:03 am

Re: Skip with Aspect ratio only

Post by kihovuti » Wed Dec 13, 2017 11:47 am

erik wrote:Did you enable cut_on_ar_change?
Yes, It was enabled in all runs. I'll post complete log after another run.

kihovuti
Posts: 7
Joined: Tue Dec 12, 2017 11:03 am

Re: Skip with Aspect ratio only

Post by kihovuti » Wed Dec 13, 2017 12:10 pm

Code: Select all

################################################################
Starting second pass using d:\atv.logo.txt
Time at start of second run:
Wed Dec 13 15:48:21 2017
################################################################
Mpeg:	d:\atv.ts
Exe	D:\Comskip\comskip.exe
Logo:	d:\atv.logo.txt
Ini:	D:\Comskip\comskip.ini

Detection Methods to be used:
	1) Logo - Give up after 2000 seconds
	2) Aspect Ratio


ComSkip throttles back from -0001 to -0001.
The time is now 1548 so it's full speed ahead!

Settings
--------
[Main Settings]
;the sum of the values for which kind of frames comskip will consider as possible cutpoints: 1=uniform (black or any other color) frame, 2=logo, 4=scene change, 8=resolution change, 16=closed captions, 32=aspect ration, 64=silence, 255=all.
detect_method=34
;Set to 10 to show a lot of extra info, level 5 is also OK, set to 0 to disable
verbose=10
;Frame not black if any of the pixels of the frame has a brightness greater than this (scale 0 to 255)
max_brightness=60
maxbright=1
;Frame not pure black if a small number of the pixels of the frame has a brightness greater than this. To decide if the frame is truly black, comskip will also check average brightness (scale 0 to 255)
test_brightness=40
;
max_avg_brightness=25
;
max_commercialbreak=1500
;
min_commercialbreak=25
;
max_commercial_size=180
;
min_commercial_size=4
;
min_show_segment_length=70
;
max_volume=500
;
max_silence=100
;
non_uniformity=500
[Detailed Settings]
min_silence=12
remove_silent_segments=0
noise_level=5
brightness_jump=200
fps=22
validate_silence=1
validate_uniform=1
validate_scenechange=1
global_threshold=1.05
disable_heuristics=0
cut_on_ac_change=0
[CPU Load Reduction]
thread_count=4
hardware_decode=1
play_nice_start=-1
play_nice_end=-1
play_nice_sleep=2
[Input Correction]
max_repair_size=200
ms_audio_delay=5
volume_slip=40
lowres=0
[Aspect Ratio]
ar_delta=0.08
cut_on_ar_change=1
[Global Removes]
padding=0
remove_before=0
remove_after=0
added_recording=14
delete_show_after_last_commercial=0
delete_show_before_first_commercial=0
delete_show_before_or_after_current=0
delete_block_after_commercial=0
min_commercial_break_at_start_or_end=39
always_keep_first_seconds=0
always_keep_last_seconds=0
[USA Specific]
intelligent_brightness=0
black_percentile=0.0076
uniform_percentile=0.003
score_percentile=0.71
[Main Scoring]
length_strict_modifier=3
length_nonstrict_modifier=1.5
combined_length_strict_modifier=2
combined_length_nonstrict_modifier=1.25
ar_wrong_modifier=2
ac_wrong_modifier=1
excessive_length_modifier=0.01
dark_block_modifier=0.3
min_schange_modifier=0.5
max_schange_modifier=2
logo_present_modifier=0.01
punish_no_logo=1
[Detailed Scoring]
punish=0
reward=0
punish_threshold=1.3
punish_modifier=2
reward_modifier=0.5
[Logo Finding]
border=10
give_up_logo_search=2000
delay_logo_search=0
logo_max_percentage_of_screen=0.12
ticker_tape=0
ticker_tape_percentage=0
ignore_side=0
ignore_left_side=0
ignore_right_side=0
subtitles=0
logo_at_bottom=0
logo_threshold=0.75
logo_percentage_threshold=0.25
logo_filter=0
aggressive_logo_rejection=0
edge_level_threshold=5
edge_radius=2
edge_weight=10
edge_step=1
num_logo_buffers=50
use_existing_logo_file=1
two_pass_logo=1
[Logo Interpretation]
connect_blocks_with_logo=0
logo_percentile=0.92
logo_fraction=0.4
shrink_logo=5
shrink_logo_tail=0
before_logo=0
after_logo=0
where_logo=0
min_black_frames_for_break=1
[Closed Captioning]
ccCheck=0
cc_commercial_type_modifier=4
cc_wrong_type_modifier=2
cc_correct_type_modifier=0.75
[Live TV]
live_tv=0
live_tv_retries=4
require_div5=0
div5_tolerance=-1
incommercial_frames=1000
[Output Control]
output_default=1
output_chapters=0
output_plist_cutlist=0
output_zoomplayer_cutlist=0
output_zoomplayer_chapter=0
output_scf=0
output_vcf=0
output_vdr=0
output_projectx=0
output_avisynth=0
output_videoredo=0
output_videoredo3=0
videoredo_offset=2
output_btv=0
output_edl=1
output_live=0
edl_offset=0
timeline_repair=1
edl_skip_field=0
output_edlp=0
output_bsplayer=0
output_edlx=0
output_cuttermaran=0
output_mpeg2schnitt=0
output_womble=0
output_mls=0
output_mpgtx=0
output_dvrmstb=0
output_dvrcut=0
output_ipodchap=0
output_framearray=0
output_debugwindow=1
output_tuning=0
output_training=0
output_false=0
output_aspect=0
output_demux=0
output_data=0
output_srt=0
output_smi=0
output_timing=0
output_incommercial=0
output_ffmeta=0
output_ffsplit=0
delete_logo_file=0
output_mkvtoolnix=0
cutscene_frame=0
cutscene_threshold=10
windowtitle="Comskip - %s"
cuttermaran_options="cut=\"true\" unattended=\"true\" muxResult=\"false\" snapToCutPoints=\"true\" closeApp=\"true\""
mpeg2schnitt_options="mpeg2schnitt.exe /S /E /R25  /Z %2 %1"
avisynth_options="LoadPlugin(\"MPEG2Dec3.dll\") \nMPEG2Source(\"%s\")\n"
dvrcut_options="dvrcut \"%s.dvr-ms\" \"%s_clean.dvr-ms\" "
[Sage Workarounds]
sage_framenumber_bug=0
sage_minute_bug=0
enable_mencoder_pts=0

The logo mask file exists.
Using d:\atv.logo.txt for logo data.
678901234567890123456789012345678901234567890123456789012345
 12:                                                             
 13:                                                             
 14:                                                             
 15:                                                             
 16:                  -----------                                
 17:               ----------------  ----------                  
 18:             ---+++++++-+++++++- -----------                 
 19:            -+++++++++---++++++++++-+++++-++                 
 20:          -++++++++---------++++++++++++++++||               
 21:          +++++++++++---+++++++++|++++++++++||               
 22:        |++++++-++++++---++++++++|||||||+++||||              
 23:       |+++++++++++++----+++++++||||++++++-+||               
 24:       |+++++++++++++--+++--+-++++||+++++++|||--------       
 25:      |+++++++++++++-+++++++++++++||++|+|+++|+---------      
 26:      |+++++++++++++++--++++++++++|+++++++|+++--------+|     
 27:      |+++++++++++++------++++++++++++++++|+ +-------++||    
 28:     ||++++++++++++-      -+++++++++++++++||+++-----+++||    
 29:     ||++++++|+++++        -++++++++++++++ ++++-----+||||    
 30:     ||++++|||+|++|          -+++|+++++-|||++||     |||||    
 31:     | ++-||||+++++          -++++++++++-|++| |   - |+|||    
 32:       +++|||++++++--    ----+++++++++++| |-+  --- -|        
 33:      |-+|||+++++++++---------++-++++++||| +|    --  --      
 34:        -||+++++++++|+-------|++++++-++++---   --- -         
 35:         | +++++++-+-+---     |-+--+++++  --+                
 36:        ||+|+++++++++----------++++-+-+-  --                 
 37:          +- ++++++++----------+++++---- - -                 
 38:         -+ ----+++++------------+-----   --                 
 39:            --------------------------   --                  
 40:               -------------------------- -- |               
 41:             -  -+---------- -------------                   
 42:          ----- --- ----------+---- ---------   ----         
 43:          ----  ---- ------------  - --------  ------        
 44:        |+++++|     ++++-+++++++-| ||++-|||++||++++++|       
 45:        |++++++|   ||++++++++|+++||||+++|||+++|++++++|||     
 46:       |++++++++   ||+++-++++ +++||||+++|||++| +++++++||     
 47:       |+++-++++|  ||||+-++++ +|+++||+|++++++| +++++++|      
 48:      |++++--+++|  ||||+-++++|||++++|++++++|+||+|+++++||     
 49:     ||+++---+++|| ||++++++++|+++++||++++++++||+++++++|||    
 50:     ||+++++++++|| ||-++++--||++++++|++++-++++++++-+++||     
 51:     |||+-+++-++||  |+++|+- ||+++++||++++++++||+++++++|      
 52:       ---------     ---      ------- -------  -------       
 53:        -     -       --       ----    -----    -----        
 54:                                                             
 55:                                                             
 56:                                                             
 57:                                                             
The last frame found in d:\atv.txt was 0
Input #0, mpegts, from 'd:\atv.ts':
  Duration: 00:08:42.31, start: 1.407833, bitrate: 248 kb/s
  Program 1 
    Metadata:
      service_name    : Service01
      service_provider: FFmpeg
    Stream #0:0[0x100]: Video: h264 (High) ([27][0][0][0] / 0x001B), yuv420p(progressive), 480x270, 25 fps, 25 tbr, 90k tbn, 50 tbc
    Stream #0:1[0x101]: Audio: aac (LC) ([15][0][0][0] / 0x000F), 48000 Hz, stereo, fltp, 36 kb/s
h.264 video can only be processed at full speed by the Donator version
h.264 video can only be processed at full speed by the Donator version
Frame Rate set to 25.000 f/s
Format changed to [480 : 270]
Frame: 1	Ratio: 2.17	MinY: 29 MaxY: 241 MinX: 1 MaxX: 480
Frame: 1 Channels:  0

Initial audio pts =      0.000
Frame: 14 Channels:  2
Frame: 2	Ratio: 2.26	MinY: 29	MaxY: 241	MinX: 1	MaxX: 480
Resizing scene change array to accomodate 2000 frames.
				Start logo cblock 0	frame 25
Strange video pts step of 0.06450 instead of 0.04050 at frame 3046
Strange video pts step of 0.05233 instead of 0.04050 at frame 3067
Strange video pts step of 0.05629 instead of 0.04050 at frame 3089
Strange video pts step of 0.05147 instead of 0.04050 at frame 3101
Strange video pts step of 0.05413 instead of 0.04050 at frame 3117
Frame: 9195	Ratio: 1.79	MinY: 1	MaxY: 270	MinX: 1	MaxX: 480
Frame: 9978	Ratio: 2.28	MinY: 30	MaxY: 242	MinX: 1	MaxX: 480
Frame: 11041	Ratio: 1.79	MinY: 1	MaxY: 270	MinX: 1	MaxX: 480
Frame: 11175	Ratio: 2.28	MinY: 30	MaxY: 241	MinX: 1	MaxX: 480
Strange audio pts step of 9.05067 instead of 0.00000 at frame 12969
Jump in base apts from 509.96000 to 519.01010, delta=9.05010

Parsed 13032 video frames and 12828 audio frames at    25.75 fps

Maximum Volume found is 19745
Finished scanning file.  Starting to build Commercial List.
WARNING: Complex timeline or errors in the recording!!!!
Results may be wrong, .ref input will be misaligned. .txt editing will produce wrong results
Use .edl output if possible
Platau@[456] frames 8, volume 2, distance 18 seconds
Platau@[476] frames 8, volume 1, distance 0 seconds
Platau@[496] frames 7, volume 2, distance 0 seconds
Platau@[5055] frames 9, volume 2, distance 183 seconds
Platau@[5075] frames 7, volume 1, distance 0 seconds
Platau@[5095] frames 8, volume 0, distance 0 seconds
Platau@[12668] frames 21, volume 0, distance 302 seconds
Vol : #Frames
  0 : 7
Calculated silence level = 10

End logo block 0	frame 12882	Length - 0:08:30.16
Logo Histogram - 0.01553
0.000 -     24 - 0.00184 *
0.050 -      0 - 0.00184 
0.100 -      0 - 0.00184 
0.150 -      0 - 0.00184 
0.200 -      0 - 0.00184 
0.250 -      0 - 0.00184 
0.300 -      0 - 0.00184 
0.350 -      0 - 0.00184 
0.400 -      0 - 0.00184 
0.450 -      0 - 0.00184 
0.500 -      0 - 0.00184 
0.550 -      0 - 0.00184 
0.600 -      0 - 0.00184 
0.650 -      0 - 0.00184 
0.700 -      0 - 0.00184 
0.750 -      0 - 0.00184 
0.800 -      0 - 0.00184 
0.850 -     75 - 0.00760 **
0.900 -  12882 - 0.99609 *********************************************************************************************************************************************************************************************************
0.950 -     50 - 0.99992 *
Set Logo Quality = 0.82500

Not enough or too much logo's found (1.00), disabling the use of Logo detection
The last ar cblock wasn't closed.  Now closing.

Printing AR cblock list before cleaning
-----------------------------------------
Block: 0	Start:      1	End:      1	AR_R: 2.17	Length: 0:00:00.00, [ 480x 270] minX=  1, minY= 29, maxX=480, maxY=241
Block: 1	Start:      2	End:   9194	AR_R: 2.26	Length: 0:06:08.56, [ 480x 270] minX=  1, minY= 29, maxX=480, maxY=241
Block: 2	Start:   9195	End:   9977	AR_R: 1.79	Length: 0:00:31.28, [ 480x 270] minX=  1, minY=  1, maxX=480, maxY=270
Block: 3	Start:   9978	End:  11040	AR_R: 2.28	Length: 0:00:42.48, [ 480x 270] minX=  1, minY= 30, maxX=480, maxY=242
Block: 4	Start:  11041	End:  11174	AR_R: 1.79	Length: 0:00:05.31, [ 480x 270] minX=  1, minY=  1, maxX=480, maxY=270
Block: 5	Start:  11175	End:  13032	AR_R: 2.28	Length: 0:01:14.24, [ 480x 270] minX=  1, minY= 30, maxX=480, maxY=241


After Sorting - 186
--------------
Aspect Ratio   2.26 found on  10311 frames totalling 	82.3%
Aspect Ratio   2.28 found on   1073 frames totalling 	90.8%
Aspect Ratio   1.79 found on    849 frames totalling 	97.6%
Aspect Ratio   2.29 found on     65 frames totalling 	98.1%
Aspect Ratio   2.24 found on     63 frames totalling 	98.6%
Aspect Ratio   2.25 found on     57 frames totalling 	99.1%
Aspect Ratio   2.21 found on     30 frames totalling 	99.3%
Aspect Ratio   2.23 found on     12 frames totalling 	99.4%
Aspect Ratio   2.22 found on     11 frames totalling 	99.5%
Aspect Ratio   2.30 found on     10 frames totalling 	99.6%
Aspect Ratio   2.17 found on      7 frames totalling 	99.6%
Aspect Ratio   2.46 found on      7 frames totalling 	99.7%
Aspect Ratio   2.13 found on      4 frames totalling 	99.7%
Aspect Ratio   2.32 found on      4 frames totalling 	99.7%
Aspect Ratio   1.74 found on      3 frames totalling 	99.8%
Aspect Ratio   2.12 found on      3 frames totalling 	99.8%
Aspect Ratio   2.14 found on      3 frames totalling 	99.8%
Aspect Ratio   2.16 found on      3 frames totalling 	99.8%
Aspect Ratio   2.38 found on      3 frames totalling 	99.9%
Aspect Ratio   2.47 found on      3 frames totalling 	99.9%
Aspect Ratio   2.15 found on      2 frames totalling 	99.9%
Aspect Ratio   2.34 found on      2 frames totalling 	99.9%
Aspect Ratio   2.35 found on      2 frames totalling 	99.9%
Aspect Ratio   2.40 found on      2 frames totalling 	99.9%
Aspect Ratio   2.09 found on      1 frames totalling 	100.0%
Aspect Ratio   2.10 found on      1 frames totalling 	100.0%
Aspect Ratio   2.20 found on      1 frames totalling 	100.0%
Aspect Ratio   2.31 found on      1 frames totalling 	100.0%
Aspect Ratio   2.36 found on      1 frames totalling 	100.0%
Aspect Ratio   2.41 found on      1 frames totalling 	100.0%
Aspect Ratio   2.45 found on      1 frames totalling 	100.0%


After Sorting - 176
--------------
Aspect Ratio   2.26 found on  12114 frames totalling 	93.0%
Aspect Ratio   1.79 found on    917 frames totalling 	100.0%
Aspect Ratio   2.17 found on      1 frames totalling 	100.0%

Printing AR cblock list
-----------------------------------------
Block: 0	Start:      1	End:      1	AR_R: 2.17	Length: 0:00:00.00, [ 480x 270] minX=  1, minY= 29, maxX=480, maxY=241
Block: 1	Start:      2	End:   9194	AR_R: 2.26	Length: 0:06:08.56, [ 480x 270] minX=  1, minY= 29, maxX=480, maxY=241
Block: 2	Start:   9195	End:   9977	AR_R: 1.79	Length: 0:00:31.28, [ 480x 270] minX=  1, minY=  1, maxX=480, maxY=270
Block: 3	Start:   9978	End:  11040	AR_R: 2.26	Length: 0:00:42.48, [ 480x 270] minX=  1, minY= 30, maxX=480, maxY=242
Block: 4	Start:  11041	End:  11174	AR_R: 1.79	Length: 0:00:05.31, [ 480x 270] minX=  1, minY=  1, maxX=480, maxY=270
Block: 5	Start:  11175	End:  13032	AR_R: 2.26	Length: 0:01:14.24, [ 480x 270] minX=  1, minY= 30, maxX=480, maxY=241
Single frames with missing audio: 2
Distribution of Black Frame   cutting:   1 positive and   0 negative, ratio is 9.9900
Confidence of Black Frame   cutting:   0 out of   0 are strict, too low
Black Frame cutting too low
Distribution of Scene Change  cutting:   0 positive and   0 negative, ratio is 9.9900
Confidence of Scene Change  cutting:   0 out of   0 are strict, too low
Distribution of Uniform Frame cutting:   0 positive and   0 negative, ratio is 9.9900
Confidence of Uniform Frame cutting:   0 out of   0 are strict, too low
Black Frame List
---------------------------
Black Frame Count = 1
nr 	frame	pts	bright	uniform	volume		cause	dimcount  bright   type
  0	 13032	 522.080	     0	     0	     0	           b 	     0	     0	P

Fuzzy scoring of the blocks
---------------------------
Block 0 has twice excess length.
Block 0 score:	Before - 1.00	After - 0.00
Threshold used - 1.0500	After rounding - 1.0500


	---------------------
	Initial Commercial List
	---------------------
No change

Logo fraction:              0.9982      Logo disabled
Maximum volume found:        19745
Average volume:               4536
Sound threshold:               500
Silence threshold:             100
Minimum volume found:           50
Average frames with silence:    90
Black threshold:                25
Minimum brightness found:       16
Minimum bright pixels found:     0
Minimum dim level found:         0
Average brightness:            751
Uniformity level:            11000
Average non uniformity:        258
Maximum gap between logo's:     -1
Suggested logo_threshold:   0.8250
Suggested shrink_logo:	    0.00
Max commercial size found:      -1
Dominant aspect ratio:      2.2600
Score threshold:            1.0500
Framerate:                  25.000
Average framerate:          24.960
Total commercial length:    0:00:00.00
Cut codes:
  F: scene	 c: change
  A: aspect	 t: cutscene
  E: exceeds	 l: logo
  L: logo	 v: volume
  B: bright	 s: scene_change
  C: combined	 a: aspect_ratio
  N: nonstrict	 u: uniform_frame
  S: strict	 b: black_frame
  		 r: resolution
----------------------------------------------------
Block list after weighing
----------------------------------------------------
  #     sbf  bs  be     fs     fe        ts        te       len     sc   scr cmb   ar                   cut    bri logo   vol sil   corr stdev   cc
  0:+     0   0   0      1  13032     0.00s   522.08s   522.08s   0.00  0.12   0 2.26            E         b   7510 0.00 45360 900  0.000    75           

13032 Frames Processed
################################################################
Time at end of run:
Wed Dec 13 16:02:05 2017
################################################################

erik
Site Admin
Posts: 3310
Joined: Sun Aug 21, 2005 3:49 pm

Re: Skip with Aspect ratio only

Post by erik » Wed Dec 13, 2017 12:28 pm

To enable hard cutting on AR changes set
cut_on_ar_change=2

This does not check if there is silence around the AR cutpoint

kihovuti
Posts: 7
Joined: Tue Dec 12, 2017 11:03 am

Re: Skip with Aspect ratio only

Post by kihovuti » Thu Dec 14, 2017 8:27 am

Thanks, its now making cuts by Aspect Ratio. Now one block is identified incorrectly part of show because it has same aspect ratio. The graphs shown on Debug Window clearly tell that this part is different from show.

1) Please tell which colored line in the graph means what? The help document only talk about a few lines while there are many in the scene.

2) And how do I tweak settings to separate out that part.

Highlighted part in the image is not part of the show, first block is correctly identified as commercial while second is identified as a show.

Image

erik
Site Admin
Posts: 3310
Joined: Sun Aug 21, 2005 3:49 pm

Re: Skip with Aspect ratio only

Post by erik » Thu Dec 14, 2017 8:35 am

During that part Comskip was unable to reliably detect the AR
You may try to set ar_delta to a lower value
ar_delta=0.04
But its difficult to see only from the screen

kihovuti
Posts: 7
Joined: Tue Dec 12, 2017 11:03 am

Re: Skip with Aspect ratio only

Post by kihovuti » Thu Dec 14, 2017 8:59 am

Problem is that AR is same as show in second part of the commercial, 2.26.

Code: Select all

----------------------------------------------------
Block list after weighing
----------------------------------------------------
  #     sbf  bs  be     fs     fe        ts        te       len     sc   scr cmb   ar                   cut    bri logo   vol sil   corr stdev   cc
  0:++    0   0   0      1   9158     0.00s   367.16s   367.16s   0.00  0.09   0 2.26            E       a     612- 0.00 4291- 910  0.000    61           
  1:--    1   1   0   9159   9941   367.20s   398.48s    31.28s   2.00  0.56   0 1.79           A        a    1821+ 0.00 5692+ 98+  0.000   182           
  2:++    1   1   0   9942  11137   398.52s   446.32s    47.80s   1.00  0.17   0 2.26                    a    1466+ 0.00 6566+ 98+  0.000   146           
  3:++    1   1   0  11138  13032   446.36s   522.08s    75.72s   0.01  0.09   0 2.26            E         b   541- 0.00 3961- 83-  0.000    54           


I guess now I need to other parameters too to correctly skip out commercial.

erik
Site Admin
Posts: 3310
Joined: Sun Aug 21, 2005 3:49 pm

Re: Skip with Aspect ratio only

Post by erik » Thu Dec 14, 2017 9:05 am

Guess you better follow the "how to ask for help" post so I have more info and it will save me some time.

kihovuti
Posts: 7
Joined: Tue Dec 12, 2017 11:03 am

Re: Skip with Aspect ratio only

Post by kihovuti » Thu Dec 14, 2017 10:14 am

Sorry, didn't know, followed it. Sent with subject "ATV Comskip data".

erik
Site Admin
Posts: 3310
Joined: Sun Aug 21, 2005 3:49 pm

Re: Skip with Aspect ratio only

Post by erik » Fri Dec 15, 2017 9:39 am

Using only AR for scoring is prone to errors as commercials can have any AR (as you see in your example)
Suggest you use:
punish=16
punish_threshold=1.3
punish_modifier=2

This will remove blocks with more than average scene changes, but you have to be sure your show has less scene changes compared to comemrcials
You also could use punish for other aspects.
From comskip.txt
The bitmask is: 1=brightness, 2=uniformity 4=volume, 8=silence amount, 16=scene change rate

Post Reply