Video Content Description (Metadata)

The Video Metadata component of play Video Intelligence provides detailed descriptions and contextual information for each video file. This metadata enhances the searchability, categorization, and analysis of video content, enabling more efficient and insightful video management. Below is an example of the metadata generated for a sample video file, along with explanations of its components.

Sample Video Metadata

{
  "context": "The scene takes place on a crowded street where a young interviewer is speaking to various people, capturing spontaneous reactions and interactions in a bustling urban setting.",
  "subjects": [
    {
      "id": 1,
      "data": {
        "age": "young_adult",
        "gender": "male",
        "attributes": [
          "blonde hair",
          "grey sweater",
          "holding microphone",
          "black pants"
        ]
      },
      "type": "person",
      "appears_in_segments": []
    },
    {
      "id": 2,
      "data": {
        "age": "middle_aged",
        "gender": "male",
        "attributes": [
          "black cap",
          "sunglasses",
          "black t-shirt"
        ]
      },
      "type": "person",
      "appears_in_segments": []
    }
  ],
  "file_name": "test_file2.mp4",
  "thumbnail": "test_file2.mp4.jpg",
  "file_metadata": {
    "codec": null,
    "delay": null,
    "width": 640,
    "format": "AVC",
    "height": 360,
    "bit_rate": 606768,
    "duration": 204080,
    "frame_rate": "25.000",
    "aspect_ratio": "1.778",
    "total_frames": 5102,
    "duration_seconds": 204.08,
    "duration_timeline": "00:03:24:0800"
  },
  "videometadata": {
    "tags": [
      "Street Interview",
      "Public Interaction",
      "Casual",
      "City Life",
      "Outdoor"
    ],
    "context": "",
    "country": "Germany",
    "summary": "",
    "filename": "test_file2.mp4",
    "lighting": "Natural Light",
    "location": "City Center",
    "lens_type": "Standard",
    "shot_type": "Two-Shot",
    "color_tone": "Natural",
    "frame_size": "Medium",
    "capturedate": "2023-10-01",
    "time_of_day": "Daytime",
    "color_palette": [
      "#dfd0ba",
      "#356564",
      "#91948b",
      "#052226",
      "#9bc3af",
      "#97b2b5"
    ],
    "weather_keywords": [
      "Sunny",
      "Clear"
    ],
    "location_keywords": [
      "Busy Street",
      "Outdoor Cafe"
    ]
  },
  "audio_segments": {
    "language": "de",
    "segments": [
      {
        "end": 23.66,
        "text": " Okay, wir haben Hilfe. Ja, lass uns... Okay, lass uns ein paar Leute holen. Was ist dein Lebenswesen? Möchtest du gerne... Ich habe meinen eigenen Kameramann. Möchtest du gerne kurz im Video sein? Was ist dein Lebenswesen? Haben Sie Lebenswesen?",
        "start": 0.009
      },
      {
        "end": 51.254,
        "text": " Nein? Hoffentlich doch. Oh mein Gott. Okay, komm schon, lass uns gehen. Muss hier eine kurze Interview sein? Was ist denn Lebenshöhle? Die Deutschen haben Lebenshöhle? Ist das korrekt? Ich bin jetzt im Nebel. Okay, komm schon. Lass uns gehen, lass uns gehen. Es ist ruhig gegangen. Muss hier eine kurze Interview sein? Haben sie Lebens...",
        "start": 23.985
      },
      {
        "end": 69.189,
        "text": " Wir haben keine Lebenssiele. Ich habe schon welche. Hast du welche? Ich komme aus England. Du musst trotzdem Lebensziele haben. Was hast du für Lebensziele? In England leben. In Deutschland leben. Das ist meine Lebenssiele. Was ist deine Lebenssiele?",
        "start": 54.07
      },
      {
        "end": 92.551,
        "text": " Ich bin schon in Deutschland, ja. Also meine Lebensziele sind 101 Jahre alt werden. Wirklich? Ja, klar. Und 100. 101. Und glücklich auch. Und glücklich auch. Und gesund auch. Und in 5 Jahren...",
        "start": 69.462
      },
      {
        "end": 109.718,
        "text": " Wo kannst du dich selbst sehen? In fünf Jahren? Wo ich bin, oder was? Wenn du da stehst, bin ich in fünf Jahren da. Also du bist in fünf Jahren da und ich komme auch her.",
        "start": 92.671
      },
      {
        "end": 126.459,
        "text": " Ich komme aus Deutschland auch. Ja, wenn du da bist, in fünf Jahren komme ich. Am besten am Samstag um 13 Uhr. In fünf Jahren. Samstag um 13 Uhr. Ja?",
        "start": 109.718
      },
      {
        "end": 148.439,
        "text": " Warum lachst du? Weil meine Deutsch ist so schlecht. Du kommst doch gar nicht. Ja, aber ich verstehe 15% und ich muss meine Deutsch noch verbessern. Wo kommst du her? I come from England. Von England. Ja, kannst du Englisch sprechen? Only a little. Oh, wow, okay.",
        "start": 127.892
      },
      {
        "end": 160.06,
        "text": " How do you do? Hau her, ich hau zurück. I am wonderful. Ich bin glücklich, dass ich in Kempten bin. Das glaube ich. Kempten ist eine schöne Stadt. Kempten ist schöner als London.",
        "start": 149.121
      },
      {
        "end": 191.203,
        "text": " Oh, Alter. Ja, ja. Du denkst? Ja, ich denke, ja. Aber Kempner hat keinen Trauer. Okay. Schade. Und keine Elisabeth, auch nicht. Schade. Hoffentlich, du liebst so 101. Ja. Und glücklich. Auch. Und viele gesund. Und viele gesund. Also, danke schon und ich wünsche dir einen schönen Tag. Danke, das wünsche ich dir auch. Ciao.",
        "start": 161.203
      },
      {
        "end": 204.087,
        "text": " Danke schön. Siehst du, du hast einen tollen Job gemacht. Danke schön. Du bist willkommen. Well done. Schau mal, ob ich es habe. Ob es auf ist. Ja, perfekt. Ich habe es gerade stoppt. Ja, du hast es gut gemacht, Mann. Okay.",
        "start": 191.203
      }
    ],
    "translation": " Okay, we've got help. Yeah, let's... Right, let's get some people. Was ist dein Lebenshieler? Möchte gerne... Got my own cameraman this time. Möchte gerne kurz ein Video sein. Was ist dein Lebenshieler? Haben Sie Lebenshieler?  No? I hope you have... Oh my god. Right, come on, let's go. Do you want to do a short interview? What's your lifestyle? Do Germans have a lifestyle? Is it correct? Am I nailing this now? Okay. Come on. Let's go, let's go. It's gone quiet. Do you want to do a short interview? Do you have a lif...  We don't have a life goal. I have some. Do you have some? I do, yeah. I come from England though. You still have to have life goals. What kind of life goals do you have? To live in England. To live in Germany. That's my life goal. You can stay. What's your life goal?  I am already in Germany. So my life goals are to become 101 years old. Really? Yes, of course. One hundred. 101. And happy too. And happy too. And healthy too. And in five years  Where can you see yourself in five years? In five years? Where I am? If you stand there, I'll be there in five years. So you'll be there in five years and I'll be there too.  I come from Germany too. Yes, if you are there, I will come in 5 years. Best on Saturday at 13 o'clock. In 5 years. Yes, Saturday at 13 o'clock. Yes?  Why are you laughing? Because my German is so bad. You don't even come here. Yes, but I understand 15% and I have to improve my German. Where are you from? I come from England. From England. Yes, can you speak English? Only a little. Oh, wow, okay.  How do you do? I am wonderful. I am happy that I am in Kempton. I think so. Kempton is a beautiful city. Kempton is more beautiful than London.  Oh, dude! Yes, yes! You think? Yes, I think, yes. And you are... But Kempner doesn't have a girlfriend. Okay. Too bad. Well, I wish you... And no Elisabeth, too. Too bad. Yes. Well, hopefully you have... You love 101. Yes. And happy. Also. And many healthy. And many healthy. Yes. Well, thank you and I wish you a nice day. Thank you, I wish you the same. Yes. Bye.  Thank you. See, you've done an amazing job. Thank you. You're welcome. Well done. Have a look whether I've got it, whether it's on. Yeah, perfect. I just stopped it, I think. Yeah, you've done good, man."
  },
  "closed_caption": "A young man is interviewing a middle-aged man in a busy street."
}

Metadata Components

Component	Description	Source
context	A brief description of the overall scene and setting of the video.	Scene understanding and contextual analysis algorithms.
subjects	An array of identified individuals in the video, including their attributes and appearances.	Computer vision and object detection models for person identification and attribute recognition.
file_name	The name of the video file.	Extracted from file metadata.
thumbnail	The filename of the video thumbnail image.	Generated from a key frame of the video.
file_metadata	Technical details about the video file, such as codec, dimensions, duration, and frame rate.	Extracted from file metadata and video analysis.
videometadata	Descriptive metadata about the video content, including tags, location, and visual characteristics.	Combination of computer vision, NLP, and scene analysis algorithms.
audio_segments	Transcribed audio content with timestamps, including the original language and translation.	Speech recognition, language detection, and translation algorithms.
closed_caption	A brief description of the main action in the video, suitable for closed captioning.	Video summarization and scene description models.

This detailed metadata helps users to better understand and utilize their video content, making Play Video Intelligence a powerful tool for video analysis and management.

Searchable Index Video Prompting