ffmpeg - 使用FFMPEG gpu python从视频中读取帧

Question

import os
import subprocess

class FFMPEGFrames:
    def __init__(self, output):
        self.output = output

    def extract_frames(self, input, fps):
        output = input.split('/')[-1].split('.')[0]

        if not os.path.exists(self.output + output):
            os.makedirs(self.output + output)

        query = "ffmpeg -y -hwaccel cuvid -c:v h264_cuvid -resize 1024x70 -i " + input + " -vf scale_npp=format=yuv420p,hwdownload,format=yuv420p -vf fps=" + str(fps) + " -pix_fmt yuvj420p -color_range 2 -vframes 1 -y " + self.output + output + "/output%06d.jpg"
        response = subprocess.Popen(query, shell=True, stdout=subprocess.PIPE).stdout.read()
        s = str(response).encode('utf-8')

我收到了这个错误:

Impossible to convert between the formats supported by the filter 'Parsed_fps_0' and the filter 'auto_scaler_0'

Error reinitializing filters!

Failed to inject frame into filter network: Function not implemented

Error while processing the decoded data for stream #0:0

Conversion failed!

Answer 1

您可能错过了一条错误消息:

Only '-vf fps' read, ignoring remaining -vf options: Use ',' to separate filters

结合你的-vf合而为一。另外，考虑替换-pix_fmt yuvj420p使用格式过滤器，因此所有过滤都可以在一个过滤器图中，您可以准确选择过滤器图中出现的位置。