|
1
|
# FFmpeg Operations Reference |
|
2
|
|
|
3
|
Complete command reference for video editing operations in Replit. All examples use Node.js with `child_process.execFile`or`fluent-ffmpeg`. |
|
4
|
|
|
5
|
## Table of Contents |
|
6
|
|
|
7
|
1. [Trimming and Cutting](#trimming-and-cutting) |
|
8
|
2. [Concatenation](#concatenation) |
|
9
|
|
|
10
|
3. [Transitions](#transitions) |
|
11
|
4. [Text Overlays](#text-overlays) |
|
12
|
|
|
13
|
5. [Subtitles](#subtitles) |
|
14
|
6. [Audio](#audio) |
|
15
|
|
|
16
|
7. [Effects](#effects) |
|
17
|
8. [Format Conversion](#format-conversion) |
|
18
|
|
|
19
|
9. [Image Extraction](#image-extraction) |
|
20
|
10. [GIF Creation](#gif-creation) |
|
21
|
|
|
22
|
11. [Resizing and Cropping](#resizing-and-cropping) |
|
23
|
12. [Watermarks and Overlays](#watermarks-and-overlays) |
|
24
|
|
|
25
|
13. [Compression](#compression) |
|
26
|
14. [Stabilization](#stabilization) |
|
27
|
|
|
28
|
15. [Looping](#looping) |
|
29
|
16. [Rotation and Flipping](#rotation-and-flipping) |
|
30
|
|
|
31
|
17. [Split Screen](#split-screen) |
|
32
|
18. [Muting and Audio Removal](#muting-and-audio-removal) |
|
33
|
|
|
34
|
19. [Thumbnail Generation](#thumbnail-generation) |
|
35
|
20. [Watermark Removal (Crop)](#watermark-removal-crop) |
|
36
|
|
|
37
|
21. [Logo Overlay (Branding)](#logo-overlay-branding) |
|
38
|
22. [Social Media Reframing](#social-media-reframing) |
|
39
|
|
|
40
|
--- |
|
41
|
|
|
42
|
## Trimming and Cutting |
|
43
|
|
|
44
|
### Trim without re-encoding (fast, frame-accurate at keyframes) |
|
45
|
|
|
46
|
```bash |
|
47
|
|
|
48
|
ffmpeg -i input.mp4 -ss 00:00:30 -to 00:01:15 -c copy output.mp4 |
|
49
|
|
|
50
|
``` |
|
51
|
|
|
52
|
Place `-ss`before`-i`for faster seeking (input seeking), or after`-i`for frame-accurate seeking (output seeking). Using`-c copy` avoids re-encoding but may have imprecise start points at non-keyframe positions. |
|
53
|
|
|
54
|
### Trim with re-encoding (slower, frame-accurate) |
|
55
|
|
|
56
|
```bash |
|
57
|
|
|
58
|
ffmpeg -i input.mp4 -ss 00:00:30 -to 00:01:15 -c:v libx264 -c:a aac output.mp4 |
|
59
|
|
|
60
|
``` |
|
61
|
|
|
62
|
### Remove a section (keep before and after) |
|
63
|
|
|
64
|
Split into two parts and concatenate: |
|
65
|
|
|
66
|
```bash |
|
67
|
|
|
68
|
ffmpeg -i input.mp4 -t 00:00:30 -c copy part1.mp4 |
|
69
|
|
|
70
|
ffmpeg -i input.mp4 -ss 00:01:00 -c copy part2.mp4 |
|
71
|
|
|
72
|
``` |
|
73
|
|
|
74
|
Then concatenate (see Concatenation section). |
|
75
|
|
|
76
|
--- |
|
77
|
|
|
78
|
## Concatenation |
|
79
|
|
|
80
|
### Concat demuxer (same codec, resolution, frame rate) |
|
81
|
|
|
82
|
Create a text file `concat_list.txt`: |
|
83
|
|
|
84
|
```text |
|
85
|
|
|
86
|
file 'clip1.mp4' |
|
87
|
|
|
88
|
file 'clip2.mp4' |
|
89
|
|
|
90
|
file 'clip3.mp4' |
|
91
|
|
|
92
|
``` |
|
93
|
|
|
94
|
```bash |
|
95
|
|
|
96
|
ffmpeg -f concat -safe 0 -i concat_list.txt -c copy output.mp4 |
|
97
|
|
|
98
|
``` |
|
99
|
|
|
100
|
This is the fastest method because it avoids re-encoding, but all clips must share the same codec, resolution, and frame rate. |
|
101
|
|
|
102
|
### Concat filter (different codecs/resolutions) |
|
103
|
|
|
104
|
```bash |
|
105
|
|
|
106
|
ffmpeg -i clip1.mp4 -i clip2.mp4 \\ |
|
107
|
|
|
108
|
-filter_complex "[0:v][0:a][1:v][1:a]concat=n=2:v=1:a=1[outv][outa]" \\ |
|
109
|
|
|
110
|
-map "[outv]" -map "[outa]" output.mp4 |
|
111
|
|
|
112
|
``` |
|
113
|
|
|
114
|
This re-encodes everything but handles mismatched inputs. Scale inputs to a common resolution first if they differ. |
|
115
|
|
|
116
|
--- |
|
117
|
|
|
118
|
## Transitions |
|
119
|
|
|
120
|
### Crossfade between two clips |
|
121
|
|
|
122
|
```bash |
|
123
|
|
|
124
|
ffmpeg -i clip1.mp4 -i clip2.mp4 \\ |
|
125
|
|
|
126
|
-filter_complex "\\ |
|
127
|
|
|
128
|
[0:v]trim=duration=5,setpts=PTS-STARTPTS[v0]; \\ |
|
129
|
|
|
130
|
[1:v]trim=duration=5,setpts=PTS-STARTPTS[v1]; \\ |
|
131
|
|
|
132
|
[v0][v1]xfade=transition=fade:duration=1:offset=4[outv]; \\ |
|
133
|
|
|
134
|
[0:a]atrim=duration=5,asetpts=PTS-STARTPTS[a0]; \\ |
|
135
|
|
|
136
|
[1:a]atrim=duration=5,asetpts=PTS-STARTPTS[a1]; \\ |
|
137
|
|
|
138
|
[a0][a1]acrossfade=d=1[outa]" \\ |
|
139
|
|
|
140
|
-map "[outv]" -map "[outa]" output.mp4 |
|
141
|
|
|
142
|
``` |
|
143
|
|
|
144
|
### Available xfade transitions |
|
145
|
|
|
146
|
`fade`, `wipeleft`, `wiperight`, `wipeup`, `wipedown`, `slideleft`, `slideright`, `slideup`, `slidedown`, `circlecrop`, `rectcrop`, `distance`, `fadeblack`, `fadewhite`, `radial`, `smoothleft`, `smoothright`, `smoothup`, `smoothdown`, `circleopen`, `circleclose`, `vertopen`, `vertclose`, `horzopen`, `horzclose`, `dissolve`, `pixelize`, `diagtl`, `diagtr`, `diagbl`, `diagbr`, `hlslice`, `hrslice`, `vuslice`, `vdslice`, `hblur`, `fadegrays`, `squeezev`, `squeezeh`, `zoomin`, `hlwind`, `hrwind`, `vuwind`, `vdwind`, `coverleft`, `coverright`, `coverup`, `coverdown`, `revealleft`, `revealright`, `revealup`, `revealdown` |
|
147
|
|
|
148
|
### Fade in/out |
|
149
|
|
|
150
|
```bash |
|
151
|
|
|
152
|
# Fade in first 2 seconds |
|
153
|
|
|
154
|
ffmpeg -i input.mp4 -vf "fade=t=in:st=0:d=2" -c:a copy output.mp4 |
|
155
|
|
|
156
|
# Fade out last 2 seconds (requires knowing duration) |
|
157
|
|
|
158
|
ffmpeg -i input.mp4 -vf "fade=t=out:st=8:d=2" -c:a copy output.mp4 |
|
159
|
|
|
160
|
# Audio fade in/out |
|
161
|
|
|
162
|
ffmpeg -i input.mp4 -af "afade=t=in:st=0:d=2,afade=t=out:st=8:d=2" output.mp4 |
|
163
|
|
|
164
|
``` |
|
165
|
|
|
166
|
--- |
|
167
|
|
|
168
|
## Text Overlays |
|
169
|
|
|
170
|
### Static text overlay |
|
171
|
|
|
172
|
```bash |
|
173
|
|
|
174
|
ffmpeg -i input.mp4 \\ |
|
175
|
|
|
176
|
-vf "drawtext=text='Hello World':fontsize=48:fontcolor=white:x=(w-text_w)/2:y=(h-text_h)/2:fontfile=/usr/share/fonts/truetype/dejavu/DejaVuSans-Bold.ttf" \\ |
|
177
|
|
|
178
|
output.mp4 |
|
179
|
|
|
180
|
``` |
|
181
|
|
|
182
|
### Text with background box |
|
183
|
|
|
184
|
```bash |
|
185
|
|
|
186
|
ffmpeg -i input.mp4 \\ |
|
187
|
|
|
188
|
-vf "drawtext=text='Title':fontsize=64:fontcolor=white:box=1:boxcolor=black@0.6:boxborderw=10:x=(w-text_w)/2:y=50" \\ |
|
189
|
|
|
190
|
output.mp4 |
|
191
|
|
|
192
|
``` |
|
193
|
|
|
194
|
### Timed text (appears and disappears) |
|
195
|
|
|
196
|
```bash |
|
197
|
|
|
198
|
ffmpeg -i input.mp4 \\ |
|
199
|
|
|
200
|
-vf "drawtext=text='Scene 1':fontsize=48:fontcolor=white:x=(w-text_w)/2:y=(h-text_h)/2:enable='between(t,2,5)'" \\ |
|
201
|
|
|
202
|
output.mp4 |
|
203
|
|
|
204
|
``` |
|
205
|
|
|
206
|
### Font discovery |
|
207
|
|
|
208
|
List available fonts in the Replit environment: |
|
209
|
|
|
210
|
```bash |
|
211
|
|
|
212
|
fc-list : family style | sort |
|
213
|
|
|
214
|
``` |
|
215
|
|
|
216
|
Or use a custom font file by downloading it to the project and referencing it with `fontfile=`. |
|
217
|
|
|
218
|
--- |
|
219
|
|
|
220
|
## Subtitles |
|
221
|
|
|
222
|
### Burn subtitles from SRT file |
|
223
|
|
|
224
|
```bash |
|
225
|
|
|
226
|
ffmpeg -i input.mp4 -vf "subtitles=subs.srt" output.mp4 |
|
227
|
|
|
228
|
``` |
|
229
|
|
|
230
|
### Burn subtitles with styling |
|
231
|
|
|
232
|
```bash |
|
233
|
|
|
234
|
ffmpeg -i input.mp4 \\ |
|
235
|
|
|
236
|
-vf "subtitles=subs.srt:force_style='FontSize=24,FontName=Arial,PrimaryColour=&Hffffff&,OutlineColour=&H000000&,Outline=2'" \\ |
|
237
|
|
|
238
|
output.mp4 |
|
239
|
|
|
240
|
``` |
|
241
|
|
|
242
|
### Burn ASS/SSA subtitles (preserves advanced styling) |
|
243
|
|
|
244
|
```bash |
|
245
|
|
|
246
|
ffmpeg -i input.mp4 -vf "ass=subs.ass" output.mp4 |
|
247
|
|
|
248
|
``` |
|
249
|
|
|
250
|
### Embed subtitles as a stream (soft subs, MKV only) |
|
251
|
|
|
252
|
```bash |
|
253
|
|
|
254
|
ffmpeg -i input.mp4 -i subs.srt -c copy -c:s srt output.mkv |
|
255
|
|
|
256
|
``` |
|
257
|
|
|
258
|
--- |
|
259
|
|
|
260
|
## Audio |
|
261
|
|
|
262
|
### Add background music |
|
263
|
|
|
264
|
```bash |
|
265
|
|
|
266
|
ffmpeg -i video.mp4 -i music.mp3 \\ |
|
267
|
|
|
268
|
-filter_complex "[1:a]volume=0.3[music];[0:a][music]amix=inputs=2:duration=first[outa]" \\ |
|
269
|
|
|
270
|
-map 0:v -map "[outa]" -c:v copy output.mp4 |
|
271
|
|
|
272
|
``` |
|
273
|
|
|
274
|
### Replace audio entirely |
|
275
|
|
|
276
|
```bash |
|
277
|
|
|
278
|
ffmpeg -i video.mp4 -i new_audio.mp3 \\ |
|
279
|
|
|
280
|
-map 0:v -map 1:a -c:v copy -shortest output.mp4 |
|
281
|
|
|
282
|
``` |
|
283
|
|
|
284
|
### Extract audio |
|
285
|
|
|
286
|
```bash |
|
287
|
|
|
288
|
ffmpeg -i video.mp4 -vn -c:a libmp3lame -q:a 2 output.mp3 |
|
289
|
|
|
290
|
ffmpeg -i video.mp4 -vn -c:a copy output.aac |
|
291
|
|
|
292
|
``` |
|
293
|
|
|
294
|
### Remove audio |
|
295
|
|
|
296
|
```bash |
|
297
|
|
|
298
|
ffmpeg -i input.mp4 -an -c:v copy output.mp4 |
|
299
|
|
|
300
|
``` |
|
301
|
|
|
302
|
### Adjust volume |
|
303
|
|
|
304
|
```bash |
|
305
|
|
|
306
|
# Set to 50% volume |
|
307
|
|
|
308
|
ffmpeg -i input.mp4 -af "volume=0.5" -c:v copy output.mp4 |
|
309
|
|
|
310
|
# Increase by 10dB |
|
311
|
|
|
312
|
ffmpeg -i input.mp4 -af "volume=10dB" -c:v copy output.mp4 |
|
313
|
|
|
314
|
# Normalize audio |
|
315
|
|
|
316
|
ffmpeg -i input.mp4 -af "loudnorm" -c:v copy output.mp4 |
|
317
|
|
|
318
|
``` |
|
319
|
|
|
320
|
### Audio delay/offset |
|
321
|
|
|
322
|
```bash |
|
323
|
|
|
324
|
# Delay audio by 1 second |
|
325
|
|
|
326
|
ffmpeg -i input.mp4 -itsoffset 1 -i input.mp4 \\ |
|
327
|
|
|
328
|
-map 0:v -map 1:a -c copy output.mp4 |
|
329
|
|
|
330
|
``` |
|
331
|
|
|
332
|
--- |
|
333
|
|
|
334
|
## Effects |
|
335
|
|
|
336
|
### Speed change |
|
337
|
|
|
338
|
```bash |
|
339
|
|
|
340
|
# 2x speed |
|
341
|
|
|
342
|
ffmpeg -i input.mp4 -filter_complex "[0:v]setpts=0.5*PTS[v];[0:a]atempo=2.0[a]" \\ |
|
343
|
|
|
344
|
-map "[v]" -map "[a]" output.mp4 |
|
345
|
|
|
346
|
# 0.5x speed (slow motion) |
|
347
|
|
|
348
|
ffmpeg -i input.mp4 -filter_complex "[0:v]setpts=2.0*PTS[v];[0:a]atempo=0.5[a]" \\ |
|
349
|
|
|
350
|
-map "[v]" -map "[a]" output.mp4 |
|
351
|
|
|
352
|
``` |
|
353
|
|
|
354
|
Note: `atempo`only accepts values between 0.5 and 100.0. For extreme slow-motion, chain multiple atempo filters:`atempo=0.5,atempo=0.5` for 0.25x. |
|
355
|
|
|
356
|
### Reverse |
|
357
|
|
|
358
|
```bash |
|
359
|
|
|
360
|
ffmpeg -i input.mp4 -vf reverse -af areverse output.mp4 |
|
361
|
|
|
362
|
``` |
|
363
|
|
|
364
|
Warning: `reverse` loads the entire video into memory. Only use on short clips. |
|
365
|
|
|
366
|
### Rotate |
|
367
|
|
|
368
|
```bash |
|
369
|
|
|
370
|
# 90 degrees clockwise |
|
371
|
|
|
372
|
ffmpeg -i input.mp4 -vf "transpose=1" output.mp4 |
|
373
|
|
|
374
|
# 90 degrees counter-clockwise |
|
375
|
|
|
376
|
ffmpeg -i input.mp4 -vf "transpose=2" output.mp4 |
|
377
|
|
|
378
|
# 180 degrees |
|
379
|
|
|
380
|
ffmpeg -i input.mp4 -vf "transpose=1,transpose=1" output.mp4 |
|
381
|
|
|
382
|
# Arbitrary angle (in radians) |
|
383
|
|
|
384
|
ffmpeg -i input.mp4 -vf "rotate=PI/6" output.mp4 |
|
385
|
|
|
386
|
``` |
|
387
|
|
|
388
|
### Color adjustments |
|
389
|
|
|
390
|
```bash |
|
391
|
|
|
392
|
# Brightness, contrast, saturation |
|
393
|
|
|
394
|
ffmpeg -i input.mp4 -vf "eq=brightness=0.1:contrast=1.2:saturation=1.3" output.mp4 |
|
395
|
|
|
396
|
# Convert to grayscale |
|
397
|
|
|
398
|
ffmpeg -i input.mp4 -vf "hue=s=0" output.mp4 |
|
399
|
|
|
400
|
# Color curves (cinematic look) |
|
401
|
|
|
402
|
ffmpeg -i input.mp4 -vf "curves=preset=cross_process" output.mp4 |
|
403
|
|
|
404
|
``` |
|
405
|
|
|
406
|
Available curve presets: `none`, `color_negative`, `cross_process`, `darker`, `increase_contrast`, `lighter`, `linear_contrast`, `medium_contrast`, `negative`, `strong_contrast`, `vintage` |
|
407
|
|
|
408
|
### Blur |
|
409
|
|
|
410
|
```bash |
|
411
|
|
|
412
|
# Box blur |
|
413
|
|
|
414
|
ffmpeg -i input.mp4 -vf "boxblur=5:1" output.mp4 |
|
415
|
|
|
416
|
# Gaussian blur |
|
417
|
|
|
418
|
ffmpeg -i input.mp4 -vf "gblur=sigma=10" output.mp4 |
|
419
|
|
|
420
|
``` |
|
421
|
|
|
422
|
### Picture-in-picture |
|
423
|
|
|
424
|
```bash |
|
425
|
|
|
426
|
ffmpeg -i main.mp4 -i overlay.mp4 \\ |
|
427
|
|
|
428
|
-filter_complex "[1:v]scale=320:240[pip];[0:v][pip]overlay=W-w-10:H-h-10" \\ |
|
429
|
|
|
430
|
output.mp4 |
|
431
|
|
|
432
|
``` |
|
433
|
|
|
434
|
### Stabilize shaky video |
|
435
|
|
|
436
|
```bash |
|
437
|
|
|
438
|
# Step 1: Analyze |
|
439
|
|
|
440
|
ffmpeg -i input.mp4 -vf "vidstabdetect=shakiness=5:accuracy=15" -f null - |
|
441
|
|
|
442
|
# Step 2: Apply stabilization |
|
443
|
|
|
444
|
ffmpeg -i input.mp4 -vf "vidstabtransform=smoothing=10:input=transforms.trf" output.mp4 |
|
445
|
|
|
446
|
``` |
|
447
|
|
|
448
|
--- |
|
449
|
|
|
450
|
## Format Conversion |
|
451
|
|
|
452
|
### MP4 (H.264 + AAC — most compatible) |
|
453
|
|
|
454
|
```bash |
|
455
|
|
|
456
|
ffmpeg -i input.webm -c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k output.mp4 |
|
457
|
|
|
458
|
``` |
|
459
|
|
|
460
|
CRF values: 18 (high quality, large file) to 28 (lower quality, small file). 23 is a good default. |
|
461
|
|
|
462
|
### WebM (VP9 + Opus — web-optimized) |
|
463
|
|
|
464
|
```bash |
|
465
|
|
|
466
|
ffmpeg -i input.mp4 -c:v libvpx-vp9 -crf 30 -b:v 0 -c:a libopus output.webm |
|
467
|
|
|
468
|
``` |
|
469
|
|
|
470
|
### MOV (ProRes — editing friendly) |
|
471
|
|
|
472
|
```bash |
|
473
|
|
|
474
|
ffmpeg -i input.mp4 -c:v prores_ks -profile:v 3 -c:a pcm_s16le output.mov |
|
475
|
|
|
476
|
``` |
|
477
|
|
|
478
|
### Audio only |
|
479
|
|
|
480
|
```bash |
|
481
|
|
|
482
|
ffmpeg -i input.mp4 -vn -c:a libmp3lame -q:a 2 output.mp3 |
|
483
|
|
|
484
|
ffmpeg -i input.mp4 -vn -c:a libopus -b:a 128k output.opus |
|
485
|
|
|
486
|
ffmpeg -i input.mp4 -vn -c:a flac output.flac |
|
487
|
|
|
488
|
``` |
|
489
|
|
|
490
|
--- |
|
491
|
|
|
492
|
## Image Extraction |
|
493
|
|
|
494
|
### Extract a single frame |
|
495
|
|
|
496
|
```bash |
|
497
|
|
|
498
|
ffmpeg -i input.mp4 -ss 00:00:05 -frames:v 1 frame.png |
|
499
|
|
|
500
|
``` |
|
501
|
|
|
502
|
### Extract frames at interval |
|
503
|
|
|
504
|
```bash |
|
505
|
|
|
506
|
# One frame per second |
|
507
|
|
|
508
|
ffmpeg -i input.mp4 -vf "fps=1" frames/frame_%04d.png |
|
509
|
|
|
510
|
# One frame every 10 seconds |
|
511
|
|
|
512
|
ffmpeg -i input.mp4 -vf "fps=1/10" frames/frame_%04d.png |
|
513
|
|
|
514
|
``` |
|
515
|
|
|
516
|
### Create a thumbnail sheet (contact sheet) |
|
517
|
|
|
518
|
```bash |
|
519
|
|
|
520
|
ffmpeg -i input.mp4 -frames:v 1 -vf "select=not(mod(n\\100)),scale=160:90,tile=5x4" thumbnail.png |
|
521
|
|
|
522
|
``` |
|
523
|
|
|
524
|
--- |
|
525
|
|
|
526
|
## GIF Creation |
|
527
|
|
|
528
|
### Basic GIF |
|
529
|
|
|
530
|
```bash |
|
531
|
|
|
532
|
ffmpeg -i input.mp4 -ss 00:00:02 -t 5 -vf "fps=15,scale=480:-1" output.gif |
|
533
|
|
|
534
|
``` |
|
535
|
|
|
536
|
### High-quality GIF with palette |
|
537
|
|
|
538
|
```bash |
|
539
|
|
|
540
|
# Step 1: Generate palette |
|
541
|
|
|
542
|
ffmpeg -i input.mp4 -ss 2 -t 5 -vf "fps=15,scale=480:-1:flags=lanczos,palettegen" palette.png |
|
543
|
|
|
544
|
# Step 2: Use palette for GIF |
|
545
|
|
|
546
|
ffmpeg -i input.mp4 -i palette.png -ss 2 -t 5 \\ |
|
547
|
|
|
548
|
-filter_complex "[0:v]fps=15,scale=480:-1:flags=lanczos[v];[v][1:v]paletteuse" output.gif |
|
549
|
|
|
550
|
``` |
|
551
|
|
|
552
|
The two-pass palette method produces significantly better GIFs with less banding. |
|
553
|
|
|
554
|
--- |
|
555
|
|
|
556
|
## Resizing and Cropping |
|
557
|
|
|
558
|
### Scale to specific dimensions |
|
559
|
|
|
560
|
```bash |
|
561
|
|
|
562
|
# Exact size |
|
563
|
|
|
564
|
ffmpeg -i input.mp4 -vf "scale=1280:720" -c:a copy output.mp4 |
|
565
|
|
|
566
|
# Scale width, auto height (maintain aspect ratio) |
|
567
|
|
|
568
|
ffmpeg -i input.mp4 -vf "scale=1280:-2" -c:a copy output.mp4 |
|
569
|
|
|
570
|
# Scale to fit within bounds (maintain aspect ratio) |
|
571
|
|
|
572
|
ffmpeg -i input.mp4 -vf "scale='min(1280,iw)':'min(720,ih)':force_original_aspect_ratio=decrease" output.mp4 |
|
573
|
|
|
574
|
``` |
|
575
|
|
|
576
|
### Crop |
|
577
|
|
|
578
|
```bash |
|
579
|
|
|
580
|
# Crop to 1280x720 from center |
|
581
|
|
|
582
|
ffmpeg -i input.mp4 -vf "crop=1280:720" -c:a copy output.mp4 |
|
583
|
|
|
584
|
# Crop with offset (x=100, y=50) |
|
585
|
|
|
586
|
ffmpeg -i input.mp4 -vf "crop=1280:720:100:50" -c:a copy output.mp4 |
|
587
|
|
|
588
|
# Crop to 16:9 from center |
|
589
|
|
|
590
|
ffmpeg -i input.mp4 -vf "crop=ih*16/9:ih" -c:a copy output.mp4 |
|
591
|
|
|
592
|
``` |
|
593
|
|
|
594
|
### Add padding/letterbox |
|
595
|
|
|
596
|
```bash |
|
597
|
|
|
598
|
# Add black bars to make 16:9 |
|
599
|
|
|
600
|
ffmpeg -i input.mp4 -vf "pad=ih*16/9:ih:(ow-iw)/2:(oh-ih)/2:black" output.mp4 |
|
601
|
|
|
602
|
``` |
|
603
|
|
|
604
|
--- |
|
605
|
|
|
606
|
## Watermarks and Overlays |
|
607
|
|
|
608
|
### Image watermark |
|
609
|
|
|
610
|
```bash |
|
611
|
|
|
612
|
# Bottom-right corner with padding |
|
613
|
|
|
614
|
ffmpeg -i video.mp4 -i watermark.png \\ |
|
615
|
|
|
616
|
-filter_complex "overlay=W-w-10:H-h-10" output.mp4 |
|
617
|
|
|
618
|
# With opacity |
|
619
|
|
|
620
|
ffmpeg -i video.mp4 -i watermark.png \\ |
|
621
|
|
|
622
|
-filter_complex "[1:v]format=rgba,colorchannelmixer=aa=0.5[wm];[0:v][wm]overlay=W-w-10:H-h-10" output.mp4 |
|
623
|
|
|
624
|
``` |
|
625
|
|
|
626
|
### Timed overlay (appears at specific time) |
|
627
|
|
|
628
|
```bash |
|
629
|
|
|
630
|
ffmpeg -i video.mp4 -i overlay.png \\ |
|
631
|
|
|
632
|
-filter_complex "overlay=10:10:enable='between(t,3,8)'" output.mp4 |
|
633
|
|
|
634
|
``` |
|
635
|
|
|
636
|
--- |
|
637
|
|
|
638
|
## Compression |
|
639
|
|
|
640
|
Reduce file size while maintaining acceptable quality. |
|
641
|
|
|
642
|
### Quick compress (reduce bitrate) |
|
643
|
|
|
644
|
```bash |
|
645
|
|
|
646
|
# Moderate compression — good balance of size vs quality |
|
647
|
|
|
648
|
ffmpeg -y -i input.mp4 \\ |
|
649
|
|
|
650
|
-c:v libx264 -crf 28 -preset medium \\ |
|
651
|
|
|
652
|
-c:a aac -b:a 96k \\ |
|
653
|
|
|
654
|
output_compressed.mp4 |
|
655
|
|
|
656
|
# Aggressive compression — much smaller file, noticeable quality loss |
|
657
|
|
|
658
|
ffmpeg -y -i input.mp4 \\ |
|
659
|
|
|
660
|
-c:v libx264 -crf 32 -preset slow \\ |
|
661
|
|
|
662
|
-c:a aac -b:a 64k \\ |
|
663
|
|
|
664
|
output_small.mp4 |
|
665
|
|
|
666
|
# Target a specific file size (e.g., 25MB for Discord/email) |
|
667
|
|
|
668
|
# Calculate bitrate: bitrate = (target_size_MB * 8192) / duration_seconds |
|
669
|
|
|
670
|
# For a 60s video targeting 25MB: bitrate = (25 * 8192) / 60 ≈ 3413 kbps |
|
671
|
|
|
672
|
ffmpeg -y -i input.mp4 \\ |
|
673
|
|
|
674
|
-c:v libx264 -b:v 3000k -maxrate 3400k -bufsize 6800k -preset medium \\ |
|
675
|
|
|
676
|
-c:a aac -b:a 128k \\ |
|
677
|
|
|
678
|
output_25mb.mp4 |
|
679
|
|
|
680
|
``` |
|
681
|
|
|
682
|
### Two-pass encoding (best quality at target size) |
|
683
|
|
|
684
|
```bash |
|
685
|
|
|
686
|
ffmpeg -y -i input.mp4 \\ |
|
687
|
|
|
688
|
-c:v libx264 -b:v 2000k -preset medium -pass 1 \\ |
|
689
|
|
|
690
|
-an -f null /dev/null |
|
691
|
|
|
692
|
ffmpeg -y -i input.mp4 \\ |
|
693
|
|
|
694
|
-c:v libx264 -b:v 2000k -preset medium -pass 2 \\ |
|
695
|
|
|
696
|
-c:a aac -b:a 128k \\ |
|
697
|
|
|
698
|
output_2pass.mp4 |
|
699
|
|
|
700
|
``` |
|
701
|
|
|
702
|
### Reduce resolution for smaller file |
|
703
|
|
|
704
|
```bash |
|
705
|
|
|
706
|
# Downscale to 720p (keeps aspect ratio) |
|
707
|
|
|
708
|
ffmpeg -y -i input.mp4 \\ |
|
709
|
|
|
710
|
-vf "scale=-2:720" \\ |
|
711
|
|
|
712
|
-c:v libx264 -crf 23 -preset medium \\ |
|
713
|
|
|
714
|
-c:a aac -b:a 128k \\ |
|
715
|
|
|
716
|
output_720p.mp4 |
|
717
|
|
|
718
|
# Downscale to 480p |
|
719
|
|
|
720
|
ffmpeg -y -i input.mp4 \\ |
|
721
|
|
|
722
|
-vf "scale=-2:480" \\ |
|
723
|
|
|
724
|
-c:v libx264 -crf 23 -preset medium \\ |
|
725
|
|
|
726
|
-c:a aac -b:a 96k \\ |
|
727
|
|
|
728
|
output_480p.mp4 |
|
729
|
|
|
730
|
``` |
|
731
|
|
|
732
|
**CRF guide:** 18 = visually lossless, 23 = default (good quality), 28 = moderate compression, 32 = heavy compression, 40+ = very low quality. Each +6 roughly halves the file size. |
|
733
|
|
|
734
|
--- |
|
735
|
|
|
736
|
## Stabilization |
|
737
|
|
|
738
|
Stabilize shaky footage using FFmpeg's vidstab filter (two-pass process). |
|
739
|
|
|
740
|
### Basic stabilization |
|
741
|
|
|
742
|
```bash |
|
743
|
|
|
744
|
# Pass 1: Analyze motion (creates transforms.trf) |
|
745
|
|
|
746
|
ffmpeg -y -i shaky.mp4 \\ |
|
747
|
|
|
748
|
-vf vidstabdetect=shakiness=5:accuracy=15:result=transforms.trf \\ |
|
749
|
|
|
750
|
-f null - |
|
751
|
|
|
752
|
# Pass 2: Apply stabilization |
|
753
|
|
|
754
|
ffmpeg -y -i shaky.mp4 \\ |
|
755
|
|
|
756
|
-vf vidstabtransform=input=transforms.trf:smoothing=10:crop=black:zoom=1 \\ |
|
757
|
|
|
758
|
-c:v libx264 -crf 23 -preset medium \\ |
|
759
|
|
|
760
|
-c:a copy \\ |
|
761
|
|
|
762
|
stabilized.mp4 |
|
763
|
|
|
764
|
``` |
|
765
|
|
|
766
|
### Node.js wrapper |
|
767
|
|
|
768
|
```typescript |
|
769
|
|
|
770
|
async function stabilizeVideo(inputPath: string, outputPath: string, smoothing = 10) { |
|
771
|
|
|
772
|
const transformsFile = inputPath + '.transforms.trf'; |
|
773
|
|
|
774
|
// Pass 1: detect motion |
|
775
|
|
|
776
|
await execFileAsync('ffmpeg', [ |
|
777
|
|
|
778
|
'-y', '-i', inputPath, |
|
779
|
|
|
780
|
'-vf', `vidstabdetect=shakiness=5:accuracy=15:result=${transformsFile}`, |
|
781
|
|
|
782
|
'-f', 'null', '-' |
|
783
|
|
|
784
|
]); |
|
785
|
|
|
786
|
// Pass 2: apply stabilization |
|
787
|
|
|
788
|
await execFileAsync('ffmpeg', [ |
|
789
|
|
|
790
|
'-y', '-i', inputPath, |
|
791
|
|
|
792
|
'-vf', `vidstabtransform=input=${transformsFile}:smoothing=${smoothing}:crop=black:zoom=1`, |
|
793
|
|
|
794
|
'-c:v', 'libx264', '-crf', '23', '-preset', 'medium', |
|
795
|
|
|
796
|
'-c:a', 'copy', |
|
797
|
|
|
798
|
outputPath |
|
799
|
|
|
800
|
]); |
|
801
|
|
|
802
|
await fs.promises.unlink(transformsFile).catch(() => {}); |
|
803
|
|
|
804
|
} |
|
805
|
|
|
806
|
``` |
|
807
|
|
|
808
|
### Parameters |
|
809
|
|
|
810
|
- `shakiness`(1-10): How shaky the video is. Default 5, use 8-10 for very shaky footage. |
|
811
|
-`smoothing`(0-30): How much to smooth the camera path. Higher = smoother but may crop more. Default 10. |
|
812
|
|
|
813
|
-`zoom` (0-10): Extra zoom to hide black borders from stabilization. 1 = slight zoom, 0 = no zoom (may show borders). |
|
814
|
|
|
815
|
--- |
|
816
|
|
|
817
|
## Looping |
|
818
|
|
|
819
|
Repeat a clip multiple times. |
|
820
|
|
|
821
|
```bash |
|
822
|
|
|
823
|
# Loop a clip 3 times using stream_loop |
|
824
|
|
|
825
|
ffmpeg -y -stream_loop 2 -i input.mp4 \\ |
|
826
|
|
|
827
|
-c copy \\ |
|
828
|
|
|
829
|
output_looped.mp4 |
|
830
|
|
|
831
|
# Loop with re-encoding (more reliable for some formats) |
|
832
|
|
|
833
|
ffmpeg -y -stream_loop 2 -i input.mp4 \\ |
|
834
|
|
|
835
|
-c:v libx264 -crf 23 -preset medium \\ |
|
836
|
|
|
837
|
-c:a aac -b:a 128k \\ |
|
838
|
|
|
839
|
output_looped.mp4 |
|
840
|
|
|
841
|
``` |
|
842
|
|
|
843
|
**Note:** `-stream_loop 2`means play the input 3 times total (original + 2 loops). Use`-stream_loop 4` for 5 times, etc. |
|
844
|
|
|
845
|
### Create a boomerang effect (forward + reverse) |
|
846
|
|
|
847
|
```bash |
|
848
|
|
|
849
|
# Forward then reverse |
|
850
|
|
|
851
|
ffmpeg -y -i input.mp4 \\ |
|
852
|
|
|
853
|
-filter_complex "[0:v]reverse[rv];[0:v][rv]concat=n=2:v=1:a=0[outv]" \\ |
|
854
|
|
|
855
|
-map "[outv]" \\ |
|
856
|
|
|
857
|
-c:v libx264 -crf 23 -preset medium \\ |
|
858
|
|
|
859
|
boomerang.mp4 |
|
860
|
|
|
861
|
``` |
|
862
|
|
|
863
|
--- |
|
864
|
|
|
865
|
## Rotation and Flipping |
|
866
|
|
|
867
|
### Rotate (2) |
|
868
|
|
|
869
|
```bash |
|
870
|
|
|
871
|
# Rotate 90 degrees clockwise |
|
872
|
|
|
873
|
ffmpeg -y -i input.mp4 \\ |
|
874
|
|
|
875
|
-vf "transpose=1" \\ |
|
876
|
|
|
877
|
-c:v libx264 -crf 23 -preset medium -c:a copy \\ |
|
878
|
|
|
879
|
rotated_90.mp4 |
|
880
|
|
|
881
|
# Rotate 90 degrees counter-clockwise |
|
882
|
|
|
883
|
ffmpeg -y -i input.mp4 \\ |
|
884
|
|
|
885
|
-vf "transpose=2" \\ |
|
886
|
|
|
887
|
-c:v libx264 -crf 23 -preset medium -c:a copy \\ |
|
888
|
|
|
889
|
rotated_270.mp4 |
|
890
|
|
|
891
|
# Rotate 180 degrees |
|
892
|
|
|
893
|
ffmpeg -y -i input.mp4 \\ |
|
894
|
|
|
895
|
-vf "transpose=1,transpose=1" \\ |
|
896
|
|
|
897
|
-c:v libx264 -crf 23 -preset medium -c:a copy \\ |
|
898
|
|
|
899
|
rotated_180.mp4 |
|
900
|
|
|
901
|
# Rotate by arbitrary angle (e.g., 15 degrees) — adds black borders |
|
902
|
|
|
903
|
ffmpeg -y -i input.mp4 \\ |
|
904
|
|
|
905
|
-vf "rotate=15*PI/180:c=black:ow=rotw(15*PI/180):oh=roth(15*PI/180)" \\ |
|
906
|
|
|
907
|
-c:v libx264 -crf 23 -preset medium -c:a copy \\ |
|
908
|
|
|
909
|
rotated_15deg.mp4 |
|
910
|
|
|
911
|
``` |
|
912
|
|
|
913
|
### Flip |
|
914
|
|
|
915
|
```bash |
|
916
|
|
|
917
|
# Flip horizontally (mirror) |
|
918
|
|
|
919
|
ffmpeg -y -i input.mp4 \\ |
|
920
|
|
|
921
|
-vf "hflip" \\ |
|
922
|
|
|
923
|
-c:v libx264 -crf 23 -preset medium -c:a copy \\ |
|
924
|
|
|
925
|
flipped_horizontal.mp4 |
|
926
|
|
|
927
|
# Flip vertically |
|
928
|
|
|
929
|
ffmpeg -y -i input.mp4 \\ |
|
930
|
|
|
931
|
-vf "vflip" \\ |
|
932
|
|
|
933
|
-c:v libx264 -crf 23 -preset medium -c:a copy \\ |
|
934
|
|
|
935
|
flipped_vertical.mp4 |
|
936
|
|
|
937
|
``` |
|
938
|
|
|
939
|
**`transpose` values:** 0 = 90° counter-clockwise + vertical flip, 1 = 90° clockwise, 2 = 90° counter-clockwise, 3 = 90° clockwise + vertical flip. |
|
940
|
|
|
941
|
--- |
|
942
|
|
|
943
|
## Split Screen |
|
944
|
|
|
945
|
Put two or more videos side by side or in a grid. |
|
946
|
|
|
947
|
### Two videos side by side (horizontal) |
|
948
|
|
|
949
|
```bash |
|
950
|
|
|
951
|
ffmpeg -y -i left.mp4 -i right.mp4 \\ |
|
952
|
|
|
953
|
-filter_complex "\\ |
|
954
|
|
|
955
|
[0:v]scale=640:360[left];\\ |
|
956
|
|
|
957
|
[1:v]scale=640:360[right];\\ |
|
958
|
|
|
959
|
[left][right]hstack=inputs=2[outv]" \\ |
|
960
|
|
|
961
|
-map "[outv]" -map 0:a? \\ |
|
962
|
|
|
963
|
-c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k \\ |
|
964
|
|
|
965
|
split_screen.mp4 |
|
966
|
|
|
967
|
``` |
|
968
|
|
|
969
|
### Two videos stacked (vertical) |
|
970
|
|
|
971
|
```bash |
|
972
|
|
|
973
|
ffmpeg -y -i top.mp4 -i bottom.mp4 \\ |
|
974
|
|
|
975
|
-filter_complex "\\ |
|
976
|
|
|
977
|
[0:v]scale=1080:540[top];\\ |
|
978
|
|
|
979
|
[1:v]scale=1080:540[bottom];\\ |
|
980
|
|
|
981
|
[top][bottom]vstack=inputs=2[outv]" \\ |
|
982
|
|
|
983
|
-map "[outv]" -map 0:a? \\ |
|
984
|
|
|
985
|
-c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k \\ |
|
986
|
|
|
987
|
stacked.mp4 |
|
988
|
|
|
989
|
``` |
|
990
|
|
|
991
|
### 2x2 grid (four videos) |
|
992
|
|
|
993
|
```bash |
|
994
|
|
|
995
|
ffmpeg -y -i v1.mp4 -i v2.mp4 -i v3.mp4 -i v4.mp4 \\ |
|
996
|
|
|
997
|
-filter_complex "\\ |
|
998
|
|
|
999
|
[0:v]scale=540:360[a];[1:v]scale=540:360[b];\\ |
|
1000
|
|
|
1001
|
[2:v]scale=540:360[c];[3:v]scale=540:360[d];\\ |
|
1002
|
|
|
1003
|
[a][b]hstack=inputs=2[top];\\ |
|
1004
|
|
|
1005
|
[c][d]hstack=inputs=2[bottom];\\ |
|
1006
|
|
|
1007
|
[top][bottom]vstack=inputs=2[outv]" \\ |
|
1008
|
|
|
1009
|
-map "[outv]" -map 0:a? \\ |
|
1010
|
|
|
1011
|
-c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k \\ |
|
1012
|
|
|
1013
|
grid_2x2.mp4 |
|
1014
|
|
|
1015
|
``` |
|
1016
|
|
|
1017
|
### Picture-in-picture (2) |
|
1018
|
|
|
1019
|
```bash |
|
1020
|
|
|
1021
|
# Small video in bottom-right corner (25% size) |
|
1022
|
|
|
1023
|
ffmpeg -y -i main.mp4 -i pip.mp4 \\ |
|
1024
|
|
|
1025
|
-filter_complex "\\ |
|
1026
|
|
|
1027
|
[1:v]scale=iw/4:ih/4[pip];\\ |
|
1028
|
|
|
1029
|
[0:v][pip]overlay=W-w-20:H-h-20[outv]" \\ |
|
1030
|
|
|
1031
|
-map "[outv]" -map 0:a \\ |
|
1032
|
|
|
1033
|
-c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k \\ |
|
1034
|
|
|
1035
|
pip_output.mp4 |
|
1036
|
|
|
1037
|
# PiP with rounded corners (border radius) |
|
1038
|
|
|
1039
|
ffmpeg -y -i main.mp4 -i pip.mp4 \\ |
|
1040
|
|
|
1041
|
-filter_complex "\\ |
|
1042
|
|
|
1043
|
[1:v]scale=320:180,format=yuva420p,\\ |
|
1044
|
|
|
1045
|
geq='lum=lum(X,Y):a=if(gt(abs(X-W/2),W/2-10)*gt(abs(Y-H/2),H/2-10),0,255)'[pip];\\ |
|
1046
|
|
|
1047
|
[0:v][pip]overlay=W-w-20:H-h-20[outv]" \\ |
|
1048
|
|
|
1049
|
-map "[outv]" -map 0:a \\ |
|
1050
|
|
|
1051
|
-c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k \\ |
|
1052
|
|
|
1053
|
pip_rounded.mp4 |
|
1054
|
|
|
1055
|
``` |
|
1056
|
|
|
1057
|
--- |
|
1058
|
|
|
1059
|
## Muting and Audio Removal |
|
1060
|
|
|
1061
|
```bash |
|
1062
|
|
|
1063
|
# Remove audio track entirely |
|
1064
|
|
|
1065
|
ffmpeg -y -i input.mp4 -an \\ |
|
1066
|
|
|
1067
|
-c:v copy \\ |
|
1068
|
|
|
1069
|
output_muted.mp4 |
|
1070
|
|
|
1071
|
# Replace audio with silence (keeps audio stream but silent) |
|
1072
|
|
|
1073
|
ffmpeg -y -i input.mp4 \\ |
|
1074
|
|
|
1075
|
-f lavfi -i anullsrc=channel_layout=stereo:sample_rate=44100 \\ |
|
1076
|
|
|
1077
|
-c:v copy -c:a aac -shortest \\ |
|
1078
|
|
|
1079
|
output_silent.mp4 |
|
1080
|
|
|
1081
|
# Keep only audio (extract as MP3) |
|
1082
|
|
|
1083
|
ffmpeg -y -i input.mp4 \\ |
|
1084
|
|
|
1085
|
-vn -c:a libmp3lame -q:a 2 \\ |
|
1086
|
|
|
1087
|
audio_only.mp3 |
|
1088
|
|
|
1089
|
# Keep only audio (extract as AAC/M4A) |
|
1090
|
|
|
1091
|
ffmpeg -y -i input.mp4 \\ |
|
1092
|
|
|
1093
|
-vn -c:a copy \\ |
|
1094
|
|
|
1095
|
audio_only.m4a |
|
1096
|
|
|
1097
|
``` |
|
1098
|
|
|
1099
|
--- |
|
1100
|
|
|
1101
|
## Thumbnail Generation |
|
1102
|
|
|
1103
|
Extract a single frame as an image file. |
|
1104
|
|
|
1105
|
```bash |
|
1106
|
|
|
1107
|
# Frame at a specific timestamp |
|
1108
|
|
|
1109
|
ffmpeg -y -ss 00:00:05 -i input.mp4 \\ |
|
1110
|
|
|
1111
|
-frames:v 1 -q:v 2 \\ |
|
1112
|
|
|
1113
|
thumbnail.jpg |
|
1114
|
|
|
1115
|
# Frame at 25% through the video |
|
1116
|
|
|
1117
|
# (calculate timestamp from duration first with ffprobe) |
|
1118
|
|
|
1119
|
ffmpeg -y -ss 30 -i input.mp4 \\ |
|
1120
|
|
|
1121
|
-frames:v 1 -q:v 2 \\ |
|
1122
|
|
|
1123
|
thumbnail_mid.jpg |
|
1124
|
|
|
1125
|
# Multiple thumbnails at regular intervals (e.g., every 10 seconds) |
|
1126
|
|
|
1127
|
ffmpeg -y -i input.mp4 \\ |
|
1128
|
|
|
1129
|
-vf "fps=1/10" -q:v 2 \\ |
|
1130
|
|
|
1131
|
thumbnails_%03d.jpg |
|
1132
|
|
|
1133
|
# Best quality PNG thumbnail |
|
1134
|
|
|
1135
|
ffmpeg -y -ss 00:00:05 -i input.mp4 \\ |
|
1136
|
|
|
1137
|
-frames:v 1 \\ |
|
1138
|
|
|
1139
|
thumbnail.png |
|
1140
|
|
|
1141
|
``` |
|
1142
|
|
|
1143
|
### Node.js: Smart thumbnail (pick the most interesting frame) |
|
1144
|
|
|
1145
|
```typescript |
|
1146
|
|
|
1147
|
async function generateThumbnail( |
|
1148
|
|
|
1149
|
videoPath: string, |
|
1150
|
|
|
1151
|
outputPath: string, |
|
1152
|
|
|
1153
|
timestamp?: number |
|
1154
|
|
|
1155
|
): Promise<string> { |
|
1156
|
|
|
1157
|
if (timestamp === undefined) { |
|
1158
|
|
|
1159
|
// Default to 25% through the video — usually more interesting than the first frame |
|
1160
|
|
|
1161
|
const { stdout } = await execFileAsync('ffprobe', [ |
|
1162
|
|
|
1163
|
'-v', 'quiet', '-print_format', 'json', '-show_format', videoPath |
|
1164
|
|
|
1165
|
]); |
|
1166
|
|
|
1167
|
const duration = parseFloat(JSON.parse(stdout).format.duration); |
|
1168
|
|
|
1169
|
timestamp = duration * 0.25; |
|
1170
|
|
|
1171
|
} |
|
1172
|
|
|
1173
|
await execFileAsync('ffmpeg', [ |
|
1174
|
|
|
1175
|
'-y', '-ss', timestamp.toFixed(3), '-i', videoPath, |
|
1176
|
|
|
1177
|
'-frames:v', '1', '-q:v', '2', |
|
1178
|
|
|
1179
|
outputPath |
|
1180
|
|
|
1181
|
]); |
|
1182
|
|
|
1183
|
return outputPath; |
|
1184
|
|
|
1185
|
} |
|
1186
|
|
|
1187
|
``` |
|
1188
|
|
|
1189
|
--- |
|
1190
|
|
|
1191
|
## Watermark Removal (Crop) |
|
1192
|
|
|
1193
|
Remove a watermark by cropping it out. This is the only reliable FFmpeg approach — there is no "inpainting" filter. |
|
1194
|
|
|
1195
|
```bash |
|
1196
|
|
|
1197
|
# Crop out a watermark in the bottom-right corner |
|
1198
|
|
|
1199
|
# Removes 60px from the bottom and 120px from the right |
|
1200
|
|
|
1201
|
ffmpeg -y -i input.mp4 \\ |
|
1202
|
|
|
1203
|
-vf "crop=iw-120:ih-60:0:0" \\ |
|
1204
|
|
|
1205
|
-c:v libx264 -crf 23 -preset medium -c:a copy \\ |
|
1206
|
|
|
1207
|
output_no_watermark.mp4 |
|
1208
|
|
|
1209
|
# Crop out a watermark at the top-left |
|
1210
|
|
|
1211
|
# Removes 80px from the top and 200px from the left |
|
1212
|
|
|
1213
|
ffmpeg -y -i input.mp4 \\ |
|
1214
|
|
|
1215
|
-vf "crop=iw-200:ih-80:200:80" \\ |
|
1216
|
|
|
1217
|
-c:v libx264 -crf 23 -preset medium -c:a copy \\ |
|
1218
|
|
|
1219
|
output_no_watermark.mp4 |
|
1220
|
|
|
1221
|
# Crop + scale back to original resolution (fills the gap) |
|
1222
|
|
|
1223
|
ffmpeg -y -i input.mp4 \\ |
|
1224
|
|
|
1225
|
-vf "crop=iw-120:ih-60:0:0,scale=1920:1080" \\ |
|
1226
|
|
|
1227
|
-c:v libx264 -crf 23 -preset medium -c:a copy \\ |
|
1228
|
|
|
1229
|
output_cleaned.mp4 |
|
1230
|
|
|
1231
|
``` |
|
1232
|
|
|
1233
|
**Crop filter syntax:** `crop=out_w:out_h:x:y` — output width, output height, x offset from left, y offset from top. |
|
1234
|
|
|
1235
|
**Note:** Cropping changes the aspect ratio/resolution. If the user needs the original dimensions, scale back up after cropping, though this may introduce slight softness. |
|
1236
|
|
|
1237
|
--- |
|
1238
|
|
|
1239
|
## Logo Overlay (Branding) |
|
1240
|
|
|
1241
|
Add a persistent logo/brand image to a video. |
|
1242
|
|
|
1243
|
```bash |
|
1244
|
|
|
1245
|
# Logo in the top-left corner with padding |
|
1246
|
|
|
1247
|
ffmpeg -y -i video.mp4 -i logo.png \\ |
|
1248
|
|
|
1249
|
-filter_complex "overlay=20:20" \\ |
|
1250
|
|
|
1251
|
-c:v libx264 -crf 23 -preset medium -c:a copy \\ |
|
1252
|
|
|
1253
|
branded.mp4 |
|
1254
|
|
|
1255
|
# Logo in the bottom-right corner |
|
1256
|
|
|
1257
|
ffmpeg -y -i video.mp4 -i logo.png \\ |
|
1258
|
|
|
1259
|
-filter_complex "overlay=W-w-20:H-h-20" \\ |
|
1260
|
|
|
1261
|
-c:v libx264 -crf 23 -preset medium -c:a copy \\ |
|
1262
|
|
|
1263
|
branded.mp4 |
|
1264
|
|
|
1265
|
# Logo with transparency (50% opacity) |
|
1266
|
|
|
1267
|
ffmpeg -y -i video.mp4 -i logo.png \\ |
|
1268
|
|
|
1269
|
-filter_complex "[1:v]format=rgba,colorchannelmixer=aa=0.5[logo];[0:v][logo]overlay=W-w-20:20" \\ |
|
1270
|
|
|
1271
|
-c:v libx264 -crf 23 -preset medium -c:a copy \\ |
|
1272
|
|
|
1273
|
branded_subtle.mp4 |
|
1274
|
|
|
1275
|
# Logo scaled to a specific size (e.g., 100px wide, keep aspect ratio) |
|
1276
|
|
|
1277
|
ffmpeg -y -i video.mp4 -i logo.png \\ |
|
1278
|
|
|
1279
|
-filter_complex "[1:v]scale=100:-1[logo];[0:v][logo]overlay=W-w-20:20" \\ |
|
1280
|
|
|
1281
|
-c:v libx264 -crf 23 -preset medium -c:a copy \\ |
|
1282
|
|
|
1283
|
branded_small.mp4 |
|
1284
|
|
|
1285
|
# Logo that appears only during the first 5 seconds |
|
1286
|
|
|
1287
|
ffmpeg -y -i video.mp4 -i logo.png \\ |
|
1288
|
|
|
1289
|
-filter_complex "[1:v]scale=100:-1[logo];[0:v][logo]overlay=20:20:enable='between(t,0,5)'" \\ |
|
1290
|
|
|
1291
|
-c:v libx264 -crf 23 -preset medium -c:a copy \\ |
|
1292
|
|
|
1293
|
branded_intro.mp4 |
|
1294
|
|
|
1295
|
``` |
|
1296
|
|
|
1297
|
### Overlay position shortcuts |
|
1298
|
|
|
1299
|
- Top-left: `overlay=20:20` |
|
1300
|
- Top-right: `overlay=W-w-20:20` |
|
1301
|
|
|
1302
|
- Bottom-left: `overlay=20:H-h-20` |
|
1303
|
- Bottom-right: `overlay=W-w-20:H-h-20` |
|
1304
|
|
|
1305
|
- Center: `overlay=(W-w)/2:(H-h)/2` |
|
1306
|
|
|
1307
|
--- |
|
1308
|
|
|
1309
|
## Social Media Reframing |
|
1310
|
|
|
1311
|
Reframe videos for different platforms with the correct aspect ratio and resolution. |
|
1312
|
|
|
1313
|
### Platform Reference |
|
1314
|
|
|
1315
|
| Platform | Aspect Ratio | Resolution | Duration | Codec | |
|
1316
|
|
|
1317
|
|----------|-------------|------------|----------|-------| |
|
1318
|
|
|
1319
|
| TikTok | 9:16 | 1080×1920 | 10–45s | H.264 | |
|
1320
|
|
|
1321
|
| Instagram Reels | 9:16 | 1080×1920 | 10–30s | H.264 | |
|
1322
|
|
|
1323
|
| Instagram Stories | 9:16 | 1080×1920 | 1–60s | H.264 | |
|
1324
|
|
|
1325
|
| Instagram Feed | 1:1 or 4:5 | 1080×1080 or 1080×1350 | 3–60s | H.264 | |
|
1326
|
|
|
1327
|
| YouTube Shorts | 9:16 | 1080×1920 | 15–60s | H.264 | |
|
1328
|
|
|
1329
|
| YouTube (standard) | 16:9 | 1920×1080 | any | H.264 | |
|
1330
|
|
|
1331
|
| X / Twitter | 16:9 | 1280×720 | 10–45s | H.264 | |
|
1332
|
|
|
1333
|
| X / Twitter (square) | 1:1 | 720×720 | 10–45s | H.264 | |
|
1334
|
|
|
1335
|
| Facebook Reels | 9:16 | 1080×1920 | 10–30s | H.264 | |
|
1336
|
|
|
1337
|
| Facebook Feed | 16:9 or 1:1 | 1920×1080 or 1080×1080 | any | H.264 | |
|
1338
|
|
|
1339
|
| LinkedIn | 16:9 or 1:1 | 1920×1080 or 1080×1080 | 10–60s | H.264 | |
|
1340
|
|
|
1341
|
| Pinterest | 9:16 or 2:3 | 1080×1920 or 1000×1500 | 6–60s | H.264 | |
|
1342
|
|
|
1343
|
| Snapchat Spotlight | 9:16 | 1080×1920 | 5–60s | H.264 | |
|
1344
|
|
|
1345
|
### Strategy selection by conversion type |
|
1346
|
|
|
1347
|
- **Landscape → Vertical (9:16):** Use blurred fill (Strategy 3) — looks polished on TikTok/Reels/Shorts |
|
1348
|
- **Landscape → Square (1:1):** Use blurred fill or center crop depending on framing |
|
1349
|
|
|
1350
|
- **Vertical → Landscape (16:9):** Use blurred fill to avoid large black bars |
|
1351
|
- **Same orientation:** Use center crop or scale directly |
|
1352
|
|
|
1353
|
- **4:5 portrait (Instagram Feed):** Use letterbox or blurred fill from 16:9 |
|
1354
|
|
|
1355
|
### Strategy 1: Center Crop |
|
1356
|
|
|
1357
|
Crops from the center to fill the target ratio. Best when the subject is center-framed. |
|
1358
|
|
|
1359
|
```bash |
|
1360
|
|
|
1361
|
# 16:9 landscape → 9:16 vertical (TikTok/Reels) — center crop |
|
1362
|
|
|
1363
|
ffmpeg -y -i input.mp4 \\ |
|
1364
|
|
|
1365
|
-vf "crop=ih*9/16:ih,scale=1080:1920" \\ |
|
1366
|
|
|
1367
|
-c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k \\ |
|
1368
|
|
|
1369
|
output_tiktok.mp4 |
|
1370
|
|
|
1371
|
# 16:9 landscape → 1:1 square (Instagram) — center crop |
|
1372
|
|
|
1373
|
ffmpeg -y -i input.mp4 \\ |
|
1374
|
|
|
1375
|
-vf "crop=ih:ih,scale=1080:1080" \\ |
|
1376
|
|
|
1377
|
-c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k \\ |
|
1378
|
|
|
1379
|
output_instagram.mp4 |
|
1380
|
|
|
1381
|
# 9:16 vertical → 16:9 landscape — center crop |
|
1382
|
|
|
1383
|
ffmpeg -y -i input.mp4 \\ |
|
1384
|
|
|
1385
|
-vf "crop=iw:iw*9/16,scale=1920:1080" \\ |
|
1386
|
|
|
1387
|
-c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k \\ |
|
1388
|
|
|
1389
|
output_youtube.mp4 |
|
1390
|
|
|
1391
|
# 9:16 vertical → 1:1 square — center crop |
|
1392
|
|
|
1393
|
ffmpeg -y -i input.mp4 \\ |
|
1394
|
|
|
1395
|
-vf "crop=iw:iw,scale=1080:1080" \\ |
|
1396
|
|
|
1397
|
-c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k \\ |
|
1398
|
|
|
1399
|
output_square.mp4 |
|
1400
|
|
|
1401
|
``` |
|
1402
|
|
|
1403
|
### Strategy 2: Letterbox / Pillarbox (black bars) |
|
1404
|
|
|
1405
|
Fits the entire video inside the target frame with black bars filling the gaps. No content is lost. |
|
1406
|
|
|
1407
|
```bash |
|
1408
|
|
|
1409
|
# 16:9 → 9:16 with black bars (pillarbox) |
|
1410
|
|
|
1411
|
ffmpeg -y -i input.mp4 \\ |
|
1412
|
|
|
1413
|
-vf "scale=1080:-2,pad=1080:1920:(ow-iw)/2:(oh-ih)/2:black" \\ |
|
1414
|
|
|
1415
|
-c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k \\ |
|
1416
|
|
|
1417
|
output_tiktok_letterbox.mp4 |
|
1418
|
|
|
1419
|
# 16:9 → 1:1 with black bars |
|
1420
|
|
|
1421
|
ffmpeg -y -i input.mp4 \\ |
|
1422
|
|
|
1423
|
-vf "scale=1080:-2,pad=1080:1080:(ow-iw)/2:(oh-ih)/2:black" \\ |
|
1424
|
|
|
1425
|
-c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k \\ |
|
1426
|
|
|
1427
|
output_square_letterbox.mp4 |
|
1428
|
|
|
1429
|
# 9:16 → 16:9 with black bars (pillarbox) |
|
1430
|
|
|
1431
|
ffmpeg -y -i input.mp4 \\ |
|
1432
|
|
|
1433
|
-vf "scale=-2:1080,pad=1920:1080:(ow-iw)/2:(oh-ih)/2:black" \\ |
|
1434
|
|
|
1435
|
-c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k \\ |
|
1436
|
|
|
1437
|
output_youtube_letterbox.mp4 |
|
1438
|
|
|
1439
|
``` |
|
1440
|
|
|
1441
|
### Strategy 3: Blurred Fill (recommended for landscape → portrait) |
|
1442
|
|
|
1443
|
Uses a blurred, scaled-up version of the video as background behind the original. Looks much more polished than black bars, especially for landscape-to-portrait conversion. |
|
1444
|
|
|
1445
|
```bash |
|
1446
|
|
|
1447
|
# 16:9 → 9:16 with blurred background fill (best for TikTok/Reels) |
|
1448
|
|
|
1449
|
ffmpeg -y -i input.mp4 \\ |
|
1450
|
|
|
1451
|
-filter_complex "\\ |
|
1452
|
|
|
1453
|
[0:v]scale=1080:1920:force_original_aspect_ratio=increase,crop=1080:1920,gblur=sigma=50[bg];\\ |
|
1454
|
|
|
1455
|
[0:v]scale=1080:1920:force_original_aspect_ratio=decrease[fg];\\ |
|
1456
|
|
|
1457
|
[bg][fg]overlay=(W-w)/2:(H-h)/2[outv]" \\ |
|
1458
|
|
|
1459
|
-map "[outv]" -map 0:a? \\ |
|
1460
|
|
|
1461
|
-c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k \\ |
|
1462
|
|
|
1463
|
output_tiktok_blur.mp4 |
|
1464
|
|
|
1465
|
# 16:9 → 1:1 with blurred background fill |
|
1466
|
|
|
1467
|
ffmpeg -y -i input.mp4 \\ |
|
1468
|
|
|
1469
|
-filter_complex "\\ |
|
1470
|
|
|
1471
|
[0:v]scale=1080:1080:force_original_aspect_ratio=increase,crop=1080:1080,gblur=sigma=50[bg];\\ |
|
1472
|
|
|
1473
|
[0:v]scale=1080:1080:force_original_aspect_ratio=decrease[fg];\\ |
|
1474
|
|
|
1475
|
[bg][fg]overlay=(W-w)/2:(H-h)/2[outv]" \\ |
|
1476
|
|
|
1477
|
-map "[outv]" -map 0:a? \\ |
|
1478
|
|
|
1479
|
-c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k \\ |
|
1480
|
|
|
1481
|
output_square_blur.mp4 |
|
1482
|
|
|
1483
|
# 9:16 → 16:9 with blurred background fill |
|
1484
|
|
|
1485
|
ffmpeg -y -i input.mp4 \\ |
|
1486
|
|
|
1487
|
-filter_complex "\\ |
|
1488
|
|
|
1489
|
[0:v]scale=1920:1080:force_original_aspect_ratio=increase,crop=1920:1080,gblur=sigma=50[bg];\\ |
|
1490
|
|
|
1491
|
[0:v]scale=1920:1080:force_original_aspect_ratio=decrease[fg];\\ |
|
1492
|
|
|
1493
|
[bg][fg]overlay=(W-w)/2:(H-h)/2[outv]" \\ |
|
1494
|
|
|
1495
|
-map "[outv]" -map 0:a? \\ |
|
1496
|
|
|
1497
|
-c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k \\ |
|
1498
|
|
|
1499
|
output_youtube_blur.mp4 |
|
1500
|
|
|
1501
|
# 16:9 → 4:5 portrait with blurred fill (Instagram Feed — preferred over 1:1) |
|
1502
|
|
|
1503
|
ffmpeg -y -i input.mp4 \\ |
|
1504
|
|
|
1505
|
-filter_complex "\\ |
|
1506
|
|
|
1507
|
[0:v]scale=1080:1350:force_original_aspect_ratio=increase,crop=1080:1350,gblur=sigma=50[bg];\\ |
|
1508
|
|
|
1509
|
[0:v]scale=1080:1350:force_original_aspect_ratio=decrease[fg];\\ |
|
1510
|
|
|
1511
|
[bg][fg]overlay=(W-w)/2:(H-h)/2[outv]" \\ |
|
1512
|
|
|
1513
|
-map "[outv]" -map 0:a? \\ |
|
1514
|
|
|
1515
|
-c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k \\ |
|
1516
|
|
|
1517
|
output_instagram_feed_4x5.mp4 |
|
1518
|
|
|
1519
|
# 16:9 → 2:3 portrait with blurred fill (Pinterest) |
|
1520
|
|
|
1521
|
ffmpeg -y -i input.mp4 \\ |
|
1522
|
|
|
1523
|
-filter_complex "\\ |
|
1524
|
|
|
1525
|
[0:v]scale=1000:1500:force_original_aspect_ratio=increase,crop=1000:1500,gblur=sigma=50[bg];\\ |
|
1526
|
|
|
1527
|
[0:v]scale=1000:1500:force_original_aspect_ratio=decrease[fg];\\ |
|
1528
|
|
|
1529
|
[bg][fg]overlay=(W-w)/2:(H-h)/2[outv]" \\ |
|
1530
|
|
|
1531
|
-map "[outv]" -map 0:a? \\ |
|
1532
|
|
|
1533
|
-c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k \\ |
|
1534
|
|
|
1535
|
output_pinterest_blur.mp4 |
|
1536
|
|
|
1537
|
``` |
|
1538
|
|
|
1539
|
### Multi-platform Export (Node.js) |
|
1540
|
|
|
1541
|
Export to all platforms in one go: |
|
1542
|
|
|
1543
|
```typescript |
|
1544
|
|
|
1545
|
import { execFile } from 'child_process'; |
|
1546
|
|
|
1547
|
import { promisify } from 'util'; |
|
1548
|
|
|
1549
|
import * as path from 'path'; |
|
1550
|
|
|
1551
|
const execFileAsync = promisify(execFile); |
|
1552
|
|
|
1553
|
interface PlatformSpec { |
|
1554
|
|
|
1555
|
name: string; |
|
1556
|
|
|
1557
|
width: number; |
|
1558
|
|
|
1559
|
height: number; |
|
1560
|
|
|
1561
|
suffix: string; |
|
1562
|
|
|
1563
|
} |
|
1564
|
|
|
1565
|
const PLATFORMS: Record<string, PlatformSpec> = { |
|
1566
|
|
|
1567
|
tiktok: { name: 'TikTok', width: 1080, height: 1920, suffix: 'tiktok' }, |
|
1568
|
|
|
1569
|
instagram_reels: { name: 'Instagram Reels', width: 1080, height: 1920, suffix: 'ig_reels' }, |
|
1570
|
|
|
1571
|
reels: { name: 'Instagram Reels', width: 1080, height: 1920, suffix: 'ig_reels' }, |
|
1572
|
|
|
1573
|
instagram_stories: { name: 'Instagram Stories', width: 1080, height: 1920, suffix: 'ig_stories' }, |
|
1574
|
|
|
1575
|
instagram_feed: { name: 'Instagram Feed', width: 1080, height: 1080, suffix: 'ig_feed' }, |
|
1576
|
|
|
1577
|
instagram_feed_45: { name: 'Instagram Feed 4:5', width: 1080, height: 1350, suffix: 'ig_feed_4x5' }, |
|
1578
|
|
|
1579
|
youtube_shorts: { name: 'YouTube Shorts', width: 1080, height: 1920, suffix: 'yt_shorts' }, |
|
1580
|
|
|
1581
|
shorts: { name: 'YouTube Shorts', width: 1080, height: 1920, suffix: 'yt_shorts' }, |
|
1582
|
|
|
1583
|
youtube: { name: 'YouTube', width: 1920, height: 1080, suffix: 'youtube' }, |
|
1584
|
|
|
1585
|
twitter: { name: 'X / Twitter', width: 1280, height: 720, suffix: 'twitter' }, |
|
1586
|
|
|
1587
|
x: { name: 'X / Twitter', width: 1280, height: 720, suffix: 'twitter' }, |
|
1588
|
|
|
1589
|
twitter_sq: { name: 'X / Twitter Square', width: 720, height: 720, suffix: 'twitter_sq' }, |
|
1590
|
|
|
1591
|
facebook_reels: { name: 'Facebook Reels', width: 1080, height: 1920, suffix: 'fb_reels' }, |
|
1592
|
|
|
1593
|
facebook_feed: { name: 'Facebook Feed', width: 1920, height: 1080, suffix: 'fb_feed' }, |
|
1594
|
|
|
1595
|
linkedin: { name: 'LinkedIn', width: 1920, height: 1080, suffix: 'linkedin' }, |
|
1596
|
|
|
1597
|
pinterest: { name: 'Pinterest', width: 1080, height: 1920, suffix: 'pinterest' }, |
|
1598
|
|
|
1599
|
pinterest_2x3: { name: 'Pinterest 2:3', width: 1000, height: 1500, suffix: 'pinterest_2x3' }, |
|
1600
|
|
|
1601
|
snapchat: { name: 'Snapchat Spotlight', width: 1080, height: 1920, suffix: 'snapchat' }, |
|
1602
|
|
|
1603
|
}; |
|
1604
|
|
|
1605
|
async function reframeForPlatform( |
|
1606
|
|
|
1607
|
inputPath: string, |
|
1608
|
|
|
1609
|
platform: PlatformSpec, |
|
1610
|
|
|
1611
|
outputDir: string, |
|
1612
|
|
|
1613
|
strategy: 'crop' | 'letterbox' | 'blur' = 'blur' |
|
1614
|
|
|
1615
|
): Promise<string> { |
|
1616
|
|
|
1617
|
const ext = path.extname(inputPath); |
|
1618
|
|
|
1619
|
const base = path.basename(inputPath, ext); |
|
1620
|
|
|
1621
|
const outputPath = path.join(outputDir, `${base}_${platform.suffix}${ext}`); |
|
1622
|
|
|
1623
|
const { width, height } = platform; |
|
1624
|
|
|
1625
|
let filterComplex: string; |
|
1626
|
|
|
1627
|
switch (strategy) { |
|
1628
|
|
|
1629
|
case 'crop': |
|
1630
|
|
|
1631
|
filterComplex = `scale=${width}:${height}:force_original_aspect_ratio=increase,crop=${width}:${height}`; |
|
1632
|
|
|
1633
|
break; |
|
1634
|
|
|
1635
|
case 'letterbox': |
|
1636
|
|
|
1637
|
filterComplex = `scale=${width}:${height}:force_original_aspect_ratio=decrease,pad=${width}:${height}:(ow-iw)/2:(oh-ih)/2:black`; |
|
1638
|
|
|
1639
|
break; |
|
1640
|
|
|
1641
|
case 'blur': |
|
1642
|
|
|
1643
|
filterComplex = [ |
|
1644
|
|
|
1645
|
`[0:v]scale=${width}:${height}:force_original_aspect_ratio=increase,crop=${width}:${height},gblur=sigma=50[bg]`, |
|
1646
|
|
|
1647
|
`[0:v]scale=${width}:${height}:force_original_aspect_ratio=decrease[fg]`, |
|
1648
|
|
|
1649
|
`[bg][fg]overlay=(W-w)/2:(H-h)/2[outv]` |
|
1650
|
|
|
1651
|
].join(';'); |
|
1652
|
|
|
1653
|
break; |
|
1654
|
|
|
1655
|
} |
|
1656
|
|
|
1657
|
const args = ['-y', '-i', inputPath]; |
|
1658
|
|
|
1659
|
if (strategy === 'blur') { |
|
1660
|
|
|
1661
|
args.push('-filter_complex', filterComplex, '-map', '[outv]', '-map', '0:a?'); |
|
1662
|
|
|
1663
|
} else { |
|
1664
|
|
|
1665
|
args.push('-vf', filterComplex); |
|
1666
|
|
|
1667
|
} |
|
1668
|
|
|
1669
|
args.push( |
|
1670
|
|
|
1671
|
'-c:v', 'libx264', '-crf', '23', '-preset', 'medium', |
|
1672
|
|
|
1673
|
'-c:a', 'aac', '-b:a', '128k', |
|
1674
|
|
|
1675
|
outputPath |
|
1676
|
|
|
1677
|
); |
|
1678
|
|
|
1679
|
await execFileAsync('ffmpeg', args); |
|
1680
|
|
|
1681
|
return outputPath; |
|
1682
|
|
|
1683
|
} |
|
1684
|
|
|
1685
|
async function exportAllPlatforms( |
|
1686
|
|
|
1687
|
inputPath: string, |
|
1688
|
|
|
1689
|
outputDir: string, |
|
1690
|
|
|
1691
|
platforms: string[] = ['youtube', 'tiktok', 'instagram', 'twitter'], |
|
1692
|
|
|
1693
|
strategy: 'crop' | 'letterbox' | 'blur' = 'blur' |
|
1694
|
|
|
1695
|
) { |
|
1696
|
|
|
1697
|
const results: Array<{ platform: string; path: string }> = []; |
|
1698
|
|
|
1699
|
for (const key of platforms) { |
|
1700
|
|
|
1701
|
const spec = PLATFORMS[key]; |
|
1702
|
|
|
1703
|
if (!spec) continue; |
|
1704
|
|
|
1705
|
console.log(`Reframing for ${spec.name} (${spec.width}x${spec.height})...`); |
|
1706
|
|
|
1707
|
const outPath = await reframeForPlatform(inputPath, spec, outputDir, strategy); |
|
1708
|
|
|
1709
|
results.push({ platform: spec.name, path: outPath }); |
|
1710
|
|
|
1711
|
console.log(` → ${outPath}`); |
|
1712
|
|
|
1713
|
} |
|
1714
|
|
|
1715
|
return results; |
|
1716
|
|
|
1717
|
} |
|
1718
|
|
|
1719
|
``` |