📸 Facebook Photos Scraper

Scrape publicly available photos from Facebook profiles and pages and get genuine, structured metadata for every image — no fabricated fields.

The scraper works in two stages, without login: it discovers photo IDs from the target's public photos page, then fetches each photo's viewer detail response and parses it as JSON. Because the detail response is parsed as real JSON (not screen-scraped with regex), every field below comes straight from Facebook's own data.

What you get per photo

Field	Description
fbid / id	The real, reproducible Facebook photo id (stable across runs).
url	Direct photo.php?fbid=... link.
imageUrl, imageWidth, imageHeight	The largest image variant with its real pixel dimensions.
imageHd	Full-resolution image URL (largest candidate, requested at scale 2).
thumbnailUrl	Smallest image variant.
imageVariants	All image variants (url + width + height), largest first.
accessibilityCaption	Facebook's native alt text (e.g. "May be an image of ..."). null if none.
caption	The photo's caption / message text. null if none.
createdTime, createdTimeISO	Upload time (epoch + ISO 8601).
ownerId, ownerName, ownerUrl	The photo owner (page / profile).
albumId, albumTitle	The album the photo belongs to, when present.
reactionCount, commentCount, shareCount	Engagement counts.
topReactions	Per-type reaction breakdown when Facebook exposes it (else null).
location	Geotag {name, latitude, longitude} when the photo is geotagged (else null).
scrapedAt	ISO timestamp of the scrape.
mediaStorageUrl	Present only when downloadMedia is on — link to the stored image file.

Honestly null (login-gated)

taggedUsers and comments (comment bodies / authors) require a logged-in Facebook session, so on anonymous runs they are emitted as null, never faked. The run summary (RUN_SUMMARY in the key-value store) reports real coverage fractions for the nullable fields.

Input

Option	Description
startUrls	Public profile/page references (URLs or handles).
mode	own (uploads/timeline — default), tagged (photos of), album (a /media/set/ URL), albums.
maxPhotos	Max photos per target (1–10000, default 20).
newerThan / olderThan	Date filters on the real created time (absolute YYYY-MM-DD or relative like 2 weeks).
downloadMedia	Download full-res image bytes to the key-value store during the run.
minPhotos, concurrentPhotos	Advanced performance controls.
proxyConfiguration	Proxy; defaults to Apify residential (Facebook blocks datacenter IPs for logged-out scraping).

Example input

{
  "startUrls": ["nasa"],
  "mode": "own",
  "maxPhotos": 20,
  "proxyConfiguration": { "useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"] }
}

Example output row

{
  "fbid": "1404942597655241",
  "id": "1404942597655241",
  "url": "https://www.facebook.com/photo.php?fbid=1404942597655241",
  "imageUrl": "https://scontent.xx.fbcdn.net/v/t39.30808-6/...n.jpg",
  "imageWidth": 2048,
  "imageHeight": 1365,
  "imageHd": "https://scontent.xx.fbcdn.net/v/t39.30808-6/...n.jpg",
  "thumbnailUrl": "https://scontent.xx.fbcdn.net/v/t39.30808-6/...s320x320.jpg",
  "accessibilityCaption": "May be an image of 2 people",
  "caption": "…",
  "createdTime": 1764165912,
  "createdTimeISO": "2025-11-26T14:05:12+00:00",
  "ownerName": "NASA",
  "reactionCount": 1234,
  "commentCount": 56,
  "shareCount": 78,
  "topReactions": [{ "type": "Like", "count": 900 }],
  "location": null,
  "scrapedAt": "2026-07-02T00:00:00+00:00"
}

Reliability & compliance

• Facebook serves a login wall to logged-out datacenter/guest IPs. If all targets hit a login wall the run fails with an explicit status message instead of returning an empty dataset. Use residential proxy (the default).
• Only publicly accessible photos are scraped. Respect Facebook's Terms of Service, copyright, and applicable law.

Billing

Pay-per-event: row_result (one per photo) and media_download (one per stored image when downloadMedia is on). Configure these in Console → Publication → Monetization.