← Cases
SingaporeReal estate2020 – ongoing

99.co

Automatic categorization of listing photos with Vision AI

99.co cut photo labeling for property listings from three to five minutes to 30 seconds with Vision AI

WorkingCompany's own claimVendor ↗

Problem

Agents on the property portal had to key in a category and metadata for each photo they uploaded. A listing has about 15 to 20 photos, and agents manage 10 to 100 listings a day.

Approach

99.co used Google Cloud's Vision AI to categorize and label each uploaded photo automatically. A photo with a television and a couch, for example, is labeled as a living room. The system also rejects photos that show a human face, because such photos tend to get lower traction.

Results

30 secondstime to categorize and label photos, down from three to five minutes, according to 99.co's vice president of engineeringCompany's own claim[1]

Takeaway

The saving falls on agents, who upload many photos a day. The company did not say how many agents use the feature.

Categories

Industry
Real estate
AI technology
Computer vision
Evidence
Company's own claim

Sources

Source summaries

[1]Google Cloud21 Dec 2020

Google Cloud's customer story describes how the property portal 99.co moved to Google Cloud and uses Vision AI to label and sort agents' listing photos automatically. A 99.co executive says the task fell from three to five minutes to within 30 seconds. The other figures on the page come from hosting, databases and build tools. The page carries no date; the Internet Archive first saved it on December 21, 2020.

Full details are in the original sources.

Added 4 Nov 2026 · Updated 4 Nov 2026
Open case page ↗