# Video Frame Extractor - YouTube, TikTok & MP4 to Images (`seemuapps/video-frame-extractor`) Actor

Extract still frames from YouTube, TikTok, Instagram, Facebook, X or MP4 links: every N seconds, evenly spaced, scene changes, keyframes or exact times.

- **URL**: https://apify.com/seemuapps/video-frame-extractor.md
- **Developed by:** [Seemu Scraping](https://apify.com/seemuapps) (community)
- **Stats:** 42 total users, 33 monthly users, 99.8% runs succeeded, 0 bookmarks
- **User rating**: No ratings yet

## Pricing

from $2.00 / 1,000 frame extracteds

This Actor is paid per event. You are not charged for the Apify platform usage, but only a fixed price for specific events.

Learn more: https://docs.apify.com/actors/running/actors-in-store.md#pay-per-event

## What's an Apify Actor?

An Actor is a serverless cloud program that runs on the Apify platform. It has two run modes.
In Batch mode, an Actor accepts a well-defined JSON input, performs an action which can take anything from a few seconds to a few hours,
and optionally produces a well-defined JSON output, datasets with results, or files in key-value store.
In Standby mode, an Actor provides a web server which can be used as a website, API, or an MCP server.

Apify vocabulary and the platform model are defined once, in the agent quickstart at https://apify.com/agents.md.

## How to integrate an Actor?

If asked about integration, you help developers integrate Actors into their projects.
You adapt to their stack and deliver integrations that are safe, well-documented, and production-ready.

Do not guess an integration path. Every one of them is in the agent quickstart at https://apify.com/agents.md: the Apify MCP server, Agent Skills with the Apify CLI, the JavaScript and Python clients, the REST API, and the account-free path for an agent with no human to sign in. It also carries the rule on stating cost before the first paid run.

For examples already wired to this Actor's own input schema, see the [API](#api) section below.

Each client library has reference documentation the quickstart does not restate: [JavaScript/TypeScript](https://docs.apify.com/api/client/js/docs.md) (`npm install apify-client`) and [Python](https://docs.apify.com/api/client/python/docs.md) (`pip install apify-client`).

# README

## Video Frame Extractor - YouTube, TikTok & MP4 to Images

Turn any video into still images. Paste a YouTube, TikTok, Instagram, Facebook or X link, or a direct MP4/WebM file, and get back sharp JPG, PNG or WebP frames: one every few seconds, a fixed number spread across the video, one per scene change, every keyframe, or the exact moments you choose.

Frames are cut straight from the video stream, so ten thumbnails from a one-hour video take seconds. You don't have to download the whole file first.

### What you get

For every frame:

- **The image itself**: JPG, PNG or WebP, saved in the run's key-value store and shown right in the dataset table
- **frameUrl**: a direct link to the image, ready for spreadsheets, scripts or other tools
- **timestampSec** and **timestamp** (`00:01:23.500`): exactly where in the video the frame comes from
- **frameIndex**, **width**, **height**, **imageFormat**, **kvKey**
- **sceneScore**: how strong the shot change was (scenes mode)
- **videoUrl**, **videoTitle**, **videoId**, **platform**

For every video, one summary row (`recordType: "video"`) with **durationSec**, **framesExtracted**, the **contactSheetUrl** grid image, **transferredMb** and **minutesScanned**.

Links that can't be processed (private, deleted, not a video) get a `recordType: "notice"` row explaining why. Notices are free.

### Extraction modes

| Mode | What it does | Good for |
|------|--------------|----------|
| **Count** | N frames spread evenly across the video | Thumbnails, previews, quick visual summaries |
| **Interval** | One frame every N seconds (e.g. every 5 s, or every 0.5 s) | Datasets, monitoring footage, step-by-step tutorials |
| **Scenes** | One frame at each shot change, with a sensitivity threshold | Storyboards, ad and trailer breakdowns, video indexing |
| **Keyframes** | The video's own keyframes (I-frames) | Fast, exact-to-the-encoder sampling |
| **Timestamps** | The exact times you list (`90`, `1:30`, `00:02:15.5`, `1m30s`) | Grabbing specific moments you already know |

Every mode respects an optional **start time** and **end time**, and **Max frames per video** caps the output.

### Use cases

- **YouTube thumbnail research**: pull frames from competitor videos to study composition and hooks
- **Training data for computer vision**: build labelled image datasets from hours of footage at a fixed frame rate
- **Storyboards and shot lists**: get one frame per scene from trailers, ads and short films
- **Content moderation and review**: scan a video in seconds with a single contact sheet instead of watching it
- **Social media repurposing**: turn TikToks and Reels into carousel images and blog illustrations
- **Video SEO and indexing**: attach timestamped preview images to chapters and search results

### How to use

1. Paste one or more links into **Video URLs**: YouTube videos and Shorts, TikTok, Instagram reels, Facebook, X, or direct video files.
2. Pick an **Extraction mode** and set its option (number of frames, interval, scene sensitivity, or your timestamps).
3. Optionally set a **Start time** / **End time**, the **Image format**, the **Max frame width** (default 1280 px; 0 keeps the original size) and turn on the **Contact sheet**.
4. Run the actor. Frames appear in the **Dataset** tab as a table with image previews. The image files are in the **Key-value store** tab.

#### Contact sheets

Turn on **Contact sheet** to also get one grid image per video, with up to 100 frames side by side and each one stamped with its timestamp. You can review a whole video at a glance or drop the grid into a report.

### Output format

Each frame row:

```json
{
  "recordType": "frame",
  "videoUrl": "https://www.youtube.com/watch?v=aqz-KE-bpKQ",
  "videoTitle": "Big Buck Bunny 60fps 4K - Official Blender Foundation Short Film",
  "videoId": "aqz-KE-bpKQ",
  "platform": "youtube",
  "mode": "scenes",
  "frameIndex": 3,
  "timestampSec": 68.2,
  "timestamp": "00:01:08.200",
  "sceneScore": 0.6596,
  "width": 1280,
  "height": 720,
  "imageFormat": "jpg",
  "kvKey": "v1-aqz-KE-bpKQ-frame-0003.jpg",
  "frameUrl": "https://api.apify.com/v2/key-value-stores/STORE_ID/records/v1-aqz-KE-bpKQ-frame-0003.jpg",
  "durationSec": 635,
  "proxyUsed": "none"
}
```

Each video summary row:

```json
{
  "recordType": "video",
  "videoUrl": "https://www.youtube.com/watch?v=aqz-KE-bpKQ",
  "videoTitle": "Big Buck Bunny 60fps 4K - Official Blender Foundation Short Film",
  "durationSec": 635,
  "framesExtracted": 30,
  "contactSheetUrl": "https://api.apify.com/v2/key-value-stores/STORE_ID/records/v1-aqz-KE-bpKQ-contact-sheet.jpg",
  "transferredMb": 14.1,
  "minutesScanned": 11,
  "proxyUsed": "none"
}
```

Filter on `recordType` to keep only frames when you export to CSV, Excel or Google Sheets.

### Pricing

You pay per result, and failed or skipped videos are free:

- **Frame extracted**: per image saved
- **Video processed**: once per video that produced frames
- **Minute scanned**: scenes mode only, per started minute of video analysed for shot changes
- **Megabyte downloaded**: only when a video has to be fetched through the residential proxy (the actor tries free routes first); never charged otherwise

To keep costs down: lower **Max frame width** (smaller sources download faster), set a **Start time** / **End time** in scenes mode, and keep **Proxy mode** on Auto.

### FAQ

**Do I need to log in or provide cookies?** No. The actor works with public videos only. Private, age-restricted and members-only videos are reported as notices.

**How accurate are the timestamps?** Frames are decoded at the exact time requested, not snapped to the nearest keyframe (except in keyframes mode, where the keyframe is the point).

**Which platforms are supported?** YouTube (including Shorts), TikTok, Instagram reels and video posts, Facebook videos, X/Twitter videos, many other video sites, and any direct link to an MP4, WebM, MOV, MKV or M3U8 stream. Live streams are not supported until they have ended.

**What's the limit per run?** Up to 100 videos per run and up to 1,000 frames per video. **Max download per video (MB)** (default 500) skips a video before it pulls an unexpectedly large amount of data.

# Actor input Schema

## `videoUrls` (type: `array`):

Links to the videos to extract frames from. Works with YouTube (videos and Shorts), TikTok, Instagram reels and posts, Facebook, X and direct video file links (MP4, WebM, MOV, M3U8). Up to 100 links per run.

## `mode` (type: `string`):

How frames are chosen. Interval: one frame every N seconds. Count: N frames spread evenly across the video. Scenes: one frame at each shot change. Keyframes: the video's own keyframes (I-frames). Timestamps: the exact times you list.

## `frameCount` (type: `integer`):

Count mode: how many frames to take. Each one comes from the middle of an equal slice of the video, which avoids black lead-in and fade-out frames.

## `intervalSeconds` (type: `number`):

Interval mode: seconds between frames, starting at the start time. 0.5 gives two frames per second.

## `sceneThreshold` (type: `number`):

Scenes mode: how different two consecutive frames must be to count as a new shot, from 0.01 to 0.99. Lower finds more scenes (0.2 catches soft cuts), higher finds only hard cuts (0.5).

## `timestamps` (type: `array`):

Timestamps mode: the times to capture. Seconds (90 or 90.5), clock format (1:30, 00:01:30.250) or 1m30s all work. Times past the end of a video are ignored.

## `startTime` (type: `string`):

Optional. Ignore everything before this time, e.g. 30, 0:30 or 1m. Applies to every mode except timestamps.

## `endTime` (type: `string`):

Optional. Ignore everything after this time, e.g. 300, 5:00 or 5m. Leave empty to run to the end.

## `outputFormat` (type: `string`):

File format of the saved frames. JPG is smallest and universal, PNG is lossless, WebP is small with high quality.

## `maxWidth` (type: `integer`):

Frames are scaled down to at most this width, keeping the aspect ratio, and never scaled up. The smallest source quality that reaches this width is used, which keeps runs fast. Set 0 for the original resolution.

## `maxFramesPerVideo` (type: `integer`):

Upper limit on frames per video in every mode. In scenes mode the strongest scene changes are kept; in keyframes mode the keyframes are sampled evenly.

## `contactSheet` (type: `boolean`):

Also save one grid image per video with all extracted frames (up to 100) side by side, each stamped with its timestamp. Its link is in the video summary row.

## `contactSheetColumns` (type: `integer`):

Number of frames per row on the contact sheet.

## `proxyMode` (type: `string`):

Auto is the default: it tries a direct connection first and only falls back to a proxy when the site blocks the request. Residential proxy traffic is billed per MB; every other route is free. Pick a single route only if you need to force one.

## `maxDownloadMbPerVideo` (type: `integer`):

Safety limit on video data fetched per video. Frame seeking usually needs well under 1 MB per frame; scenes and keyframes modes read the whole video at its lowest quality. A video that would go over is skipped free of charge. 0 means no limit.

## Actor input object example

```json
{
  "videoUrls": [
    "https://media.w3.org/2010/05/sintel/trailer.mp4"
  ],
  "mode": "count",
  "frameCount": 10,
  "intervalSeconds": 5,
  "sceneThreshold": 0.3,
  "outputFormat": "jpg",
  "maxWidth": 1280,
  "maxFramesPerVideo": 100,
  "contactSheet": false,
  "contactSheetColumns": 5,
  "proxyMode": "auto",
  "maxDownloadMbPerVideo": 500
}
```

# Actor output Schema

## `results` (type: `string`):

One row per frame (recordType frame): videoUrl, videoTitle, frameIndex, timestampSec, timestamp, sceneScore, width, height, kvKey, frameUrl. One summary row per video (recordType video) with durationSec, framesExtracted, contactSheetUrl, transferredMb and minutesScanned. Links that could not be processed get a free recordType notice row with the reason.

## `frames` (type: `string`):

The frame images and contact sheets, stored in the run's key-value store under each row's kvKey.

# API

You can run this Actor programmatically using our API. Below are code examples in JavaScript, Python, and CLI, as well as the OpenAPI specification and MCP server setup.

## JavaScript example

```javascript
import { ApifyClient } from 'apify-client';

// Initialize the ApifyClient with your Apify API token
// Replace the '<YOUR_API_TOKEN>' with your token
const client = new ApifyClient({
    token: '<YOUR_API_TOKEN>',
});

// Prepare Actor input
const input = {
    "videoUrls": [
        "https://media.w3.org/2010/05/sintel/trailer.mp4"
    ]
};

// Run the Actor and wait for it to finish
const run = await client.actor("seemuapps/video-frame-extractor").call(input);

// Fetch and print Actor results from the run's dataset (if any)
console.log('Results from dataset');
console.log(`💾 Check your data here: https://console.apify.com/storage/datasets/${run.defaultDatasetId}`);
const { items } = await client.dataset(run.defaultDatasetId).listItems();
items.forEach((item) => {
    console.dir(item);
});

// 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/js/docs

```

## Python example

```python
from apify_client import ApifyClient

# Initialize the ApifyClient with your Apify API token
# Replace '<YOUR_API_TOKEN>' with your token.
client = ApifyClient("<YOUR_API_TOKEN>")

# Prepare the Actor input
run_input = { "videoUrls": ["https://media.w3.org/2010/05/sintel/trailer.mp4"] }

# Run the Actor and wait for it to finish
run = client.actor("seemuapps/video-frame-extractor").call(run_input=run_input)

# Fetch and print Actor results from the run's dataset (if there are any)
print(f"💾 Check your data here: https://console.apify.com/storage/datasets/{run.default_dataset_id}")
for item in client.dataset(run.default_dataset_id).iterate_items():
    print(item)

# 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/python/docs/quick-start

```

## CLI example

```bash
echo '{
  "videoUrls": [
    "https://media.w3.org/2010/05/sintel/trailer.mp4"
  ]
}' |
apify call seemuapps/video-frame-extractor --silent --output-dataset

```

## MCP server setup

```json
{
    "mcpServers": {
        "apify": {
            "type": "http",
            "url": "https://mcp.apify.com/?tools=fetch-actor-details,seemuapps/video-frame-extractor"
        }
    }
}
```

The hosted server signs you in with OAuth on first connect, so no API token belongs in this config. Clients without OAuth support can send an `Authorization: Bearer <APIFY_API_TOKEN>` header instead, using a token from API & Integrations in Apify Console (https://console.apify.com/settings/integrations).

## OpenAPI specification

Download the OpenAPI definition: https://api.apify.com/v2/actors/JmGaprYwf8bcb7b27/builds/HiGO7HmMavTyOE3mb/openapi.json