Newest 'ffmpeg' Questions - Stack Overflow

http://stackoverflow.com/questions/tagged/ffmpeg

Les articles publiés sur le site

  • How to stream audio from ffserver

    22 mai 2019, par Doro

    I trying to stream 2 files - 1.mkv without audio (which streaming ok) and 2.mkv with audio encoded with Vorbis codec which i can't stream. For encoding I used 

    ffmpeg -i 2.mp4 -strict -2 -c:a vorbis ex.mkv

    And it playing ok with ffplay

    Server log: 

    Fri May 17 00:49:08 2019 Opening feed file '1.mkv' for stream 'test1-rtsp'
Fri May 17 00:49:08 2019 [matroska,webm @ 0x200746c0]Unknown entry 0x55B0
Thu Dec 14 21:35:00 1950 [h264 @ 0x2007dcc0]gray chroma
Fri May 17 00:49:08 2019 [h264 @ 0x2007dcc0]error while decoding MB 18 1, bytestream 1989
Fri May 17 00:49:08 2019 [h264 @ 0x2007dcc0]concealing 432 DC, 432 AC, 432 MV errors in I frame
Fri May 17 00:49:08 2019 Opening feed file '2.mkv' for stream 'test2-rtsp'
Fri May 17 00:49:08 2019 [matroska,webm @ 0x200746c0]Unknown entry 0x55B0
Fri May 17 00:49:08 2019 FFserver started.
Fri May 17 00:49:25 2019 [matroska,webm @ 0x20080de0]Unknown entry 0x55B0
Fri May 17 00:49:25 2019 127.0.0.1:33582 - - "PLAY test2-rtsp/streamid=0 RTP/UDP"
Fri May 17 00:49:25 2019 127.0.0.1 - - [SETUP] "rtsp://127.0.0.1:7654/test2-rtsp/ RTSP/1.0" 200 2553

    Client log:

    Bad packed header lengths (30,0,1250,2673)
[udp @ 00000236f6318500] 'circular_buffer_size' option was set but it is not supported on this build (pthread support is required)
[udp @ 00000236f63185c0] 'circular_buffer_size' option was set but it is not supported on this build (pthread support is required)
[udp @ 00000236f633dc40] 'circular_buffer_size' option was set but it is not supported on this build (pthread support is required)
[udp @ 00000236f634df00] 'circular_buffer_size' option was set but it is not supported on this build (pthread support is required)
[rtsp @ 00000236f63153c0] method SETUP failed: 503 Service Unavailable
rtsp://127.0.0.1:7654/test2-rtsp: Server returned 5XX Server Error reply

    Configure ffserver file:

    Port 8090
BindAddress 0.0.0.0
MaxHTTPConnections 2000
MaxClients 1000
MaxBandwidth 500000
CustomLog -
NoDaemon

RTSPPort 7654
RTSPBindAddress 0.0.0.0


    Format rtp
    File "1.mkv"


    Format rtp
    Strict -2
    AudioCodec vorbis
    File "2.mkv"

  • Open CV Codec FFMPEG Error fallback to use tag 0x7634706d/'mp4v'

    22 mai 2019, par Cohen

    Doing a filter recording and all is fine. The code is running, but at the end the video is not saved as MP4. I have this error:

    OpenCV: FFMPEG: tag 0x44495658/'XVID' is not supported with codec id 12 and format 'mp4 / MP4 (MPEG-4 Part 14)'
OpenCV: FFMPEG: fallback to use tag 0x7634706d/'mp4v'

    Using a MAC and the code is running correctly, but is not saving. I tried to find more details about this error, but wasn't so fortunate. I use as editor Sublime. The code run on Atom tough but is giving this error:

    OpenCV: FFMPEG: tag 0x44495658/'XVID' is not supported with codec id 12 and format 'mp4 / MP4 (MPEG-4 Part 14)'
OpenCV: FFMPEG: fallback to use tag 0x7634706d/'mp4v'
2018-05-28 15:04:25.274 Python[17483:2224774] AVF: AVAssetWriter status: Cannot create file

    ....

    import numpy as np
import cv2
import random
from utils import CFEVideoConf, image_resize
import glob
import math


cap = cv2.VideoCapture(0)

frames_per_seconds = 24
save_path='saved-media/filter.mp4'
config = CFEVideoConf(cap, filepath=save_path, res='360p')
out = cv2.VideoWriter(save_path, config.video_type, frames_per_seconds, config.dims)


def verify_alpha_channel(frame):
    try:
        frame.shape[3] # looking for the alpha channel
    except IndexError:
        frame = cv2.cvtColor(frame, cv2.COLOR_BGR2BGRA)
    return frame


def apply_hue_saturation(frame, alpha, beta):
    hsv_image = cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)
    h, s, v = cv2.split(hsv_image)
    s.fill(199)
    v.fill(255)
    hsv_image = cv2.merge([h, s, v])

    out = cv2.cvtColor(hsv_image, cv2.COLOR_HSV2BGR)
    frame = verify_alpha_channel(frame)
    out = verify_alpha_channel(out)
    cv2.addWeighted(out, 0.25, frame, 1.0, .23, frame)
    return frame


def apply_color_overlay(frame, intensity=0.5, blue=0, green=0, red=0):
    frame = verify_alpha_channel(frame)
    frame_h, frame_w, frame_c = frame.shape
    sepia_bgra = (blue, green, red, 1)
    overlay = np.full((frame_h, frame_w, 4), sepia_bgra, dtype='uint8')
    cv2.addWeighted(overlay, intensity, frame, 1.0, 0, frame)
    return frame


def apply_sepia(frame, intensity=0.5):
    frame = verify_alpha_channel(frame)
    frame_h, frame_w, frame_c = frame.shape
    sepia_bgra = (20, 66, 112, 1)
    overlay = np.full((frame_h, frame_w, 4), sepia_bgra, dtype='uint8')
    cv2.addWeighted(overlay, intensity, frame, 1.0, 0, frame)
    return frame


def alpha_blend(frame_1, frame_2, mask):
    alpha = mask/255.0 
    blended = cv2.convertScaleAbs(frame_1*(1-alpha) + frame_2*alpha)
    return blended


def apply_circle_focus_blur(frame, intensity=0.2):
    frame = verify_alpha_channel(frame)
    frame_h, frame_w, frame_c = frame.shape
    y = int(frame_h/2)
    x = int(frame_w/2)

    mask = np.zeros((frame_h, frame_w, 4), dtype='uint8')
    cv2.circle(mask, (x, y), int(y/2), (255,255,255), -1, cv2.LINE_AA)
    mask = cv2.GaussianBlur(mask, (21,21),11 )

    blured = cv2.GaussianBlur(frame, (21,21), 11)
    blended = alpha_blend(frame, blured, 255-mask)
    frame = cv2.cvtColor(blended, cv2.COLOR_BGRA2BGR)
    return frame


def portrait_mode(frame):
    cv2.imshow('frame', frame)
    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    _, mask = cv2.threshold(gray, 120,255,cv2.THRESH_BINARY)

    mask = cv2.cvtColor(mask, cv2.COLOR_GRAY2BGRA)
    blured = cv2.GaussianBlur(frame, (21,21), 11)
    blended = alpha_blend(frame, blured, mask)
    frame = cv2.cvtColor(blended, cv2.COLOR_BGRA2BGR)
    return frame


def apply_invert(frame):
    return cv2.bitwise_not(frame)

while(True):
    # Capture frame-by-frame
    ret, frame = cap.read()
    frame = cv2.cvtColor(frame, cv2.COLOR_BGR2BGRA) 
    #cv2.imshow('frame',frame)


    hue_sat = apply_hue_saturation(frame.copy(), alpha=3, beta=3)
    cv2.imshow('hue_sat', hue_sat)

    sepia = apply_sepia(frame.copy(), intensity=.8)
    cv2.imshow('sepia',sepia)

    color_overlay = apply_color_overlay(frame.copy(), intensity=.8, red=123, green=231)
    cv2.imshow('color_overlay',color_overlay)

    invert = apply_invert(frame.copy())
    cv2.imshow('invert', invert)

    blur_mask = apply_circle_focus_blur(frame.copy())
    cv2.imshow('blur_mask', blur_mask)

    portrait = portrait_mode(frame.copy())
    cv2.imshow('portrait',portrait)

    if cv2.waitKey(20) & 0xFF == ord('q'):
        break

# When everything done, release the capture
cap.release()
cv2.destroyAllWindows()

  • Overlaying a large image onto a small video like snapchat

    22 mai 2019, par Tabassum Latif

    I want to add full screen overlay image on different size of video, I used ffmpeg it work fine if overlay and video has same scale ratio (same width and height). Overlay image squeezed if video height and width size is smaller than overlay image but i want to apply like snap chat video they add black view on top and bottom

    Ffmpeg command is 

    String[]{"-y", "-i", filePath, "-i",
            overlayPath,
            "-preset", "ultrafast",
            "-filter_complex",
            "[1][0]scale2ref[i][m];[m][i]overlay[v]", "-map",
            "[v]", "-map", "0:a?", "-ac", "2", output};

    I want following output link is mention below.

    Snapchat output

    My editing result output

  • How to trim single ts file and concat with everything after with ffmpeg

    22 mai 2019, par user3321348

    I have a number of ts files (h264, AAC) and I want to quickly cut out a portion without re-encoding more than the first chunk and last chunk. Then I want to concat them back together and make an mp4.

    So I have 1.ts - 100.ts and each chunk is roughly 10 seconds. I want to cut from 1:05 to 10:05 minutes. I have code that finds the first and last ts file needed (let's say it's 11.ts for 1:05 and 61.ts for 10:05)

    I run this to trim the first ts file to only the portion I need:

    ffmpeg -i 11.ts -ss 5.0 -c:v libx264 -c:a aac 11.new.ts

    Roughly the same thing happens to 61.ts to create 61.new.ts.

    Now I want to concat 11.new.ts, 12.ts, 13.ts, ..., 61.new.ts into an mp4 without having to re-encode every single chunk. Currently, the resulting video plays the remaining portion of the first chunk just fine. After that point the audio continues but there's no more video. I'm sure that has something to do with start time offsets that's in the metadata of each ts file, but I can't figure out how to solve that. Is this even possible? And is this the best way to quickly do something like this?

  • I wanted to create video streaming website, but the problem that I want to hard sub

    21 mai 2019, par hamad alafifi

    I want to hardsub all the movies and add watermark to them, I used ffmpeg once bu it’s slow, if you can recommend new way or how to use it faster.

Navigation

Sites Web