Reduce costs
Lower bandwidth and storage expenses without compromising quality.
Ingest Video File
Send a video file to the /v1/start_encode2 endpoint and add video_intelligence as an output format in your job request.
Video Intelligence
Understand your video content with AI-powered detection and analysis.
Overview
What It Is
Video Intelligence applies computer vision and machine learning to extract rich metadata and insights from your videos automatically. The API detects objects, scenes, transitions, text overlays, and other visual elements to create searchable metadata without manual tagging. This enables content discovery, moderation, accessibility, and personalization at scale without human review.
Coming Soon
Process
How It Works
Run Detection Models
AI scans frames to identify objects, scenes, text, faces, and visual patterns.
Extract Metadata
Collect timestamps, descriptions, confidence scores, and structured data from detected elements.
Generate Insights
Compile results into searchable metadata, tags, and scene breakdowns.
Return Structured Data
Receive machine-readable insights for use in search, recommendations, moderation, and accessibility features.
Overview
Why Video Intelligence
Save time
Automate complex processing that would take hours manually.
Scale effortlessly
Handle thousands of videos with the same API call.
Integrate in minutes
Simple REST API with comprehensive SDKs and documentation.
Use Cases
Who Uses Video Intelligence
VOD Platforms
ScenarioA streaming platform with millions of videos needs searchable metadata but manual tagging is impossible. Users struggle to find content beyond basic title and description.
Auto-generate detailed metadata and visual tags for each video, enabling rich filtering and improving content discoverability by 35%.
UGC Platforms
ScenarioA user-generated content site accepts hundreds of uploads daily. They need to tag content and detect policy violations without significant moderation overhead.
Automatically label and scan videos for inappropriate content, streamlining moderation by 70% and improving content organization.
Ad Tech
ScenarioAd networks must understand video content to place ads appropriately and report brand safety metrics to advertisers.
Detect visual context and brand-safety factors automatically, enabling confident ad placement and improving brand safety reporting accuracy to 98%.
Tutorial
How to Use Video Intelligence
Tutorial Video Coming Soon
For Developers
Technical Reference
| API Method | POST /v1/start_encode2 with output: "video_intelligence" after creating a task with /v1/create_task. |
| Key Parameters | source, encoder_version: "2", output: "video_intelligence", mode. Available modes: description, categorization, moderation, custom, search. Optional mode-specific params: categories, violation_reasons, prompt, media_prompt, search_rank_threshold, search_options, result_name. |
| Supported Input Formats | Any video format — MP4, MOV, MKV, AVI, HLS, and 200+ others. Description/categorization/moderation/custom: under 2 GB, 10 seconds to 2 hours. Search: under 4 GB, 4 seconds to 4 hours. Resolution: 360x360 to 5184x2160; aspect ratio: 1:1 to 2.4:1. |
| Detection Models | Description, Categorization, Content Moderation, Custom Prompt, Search |
| Output Format | JSON result file. Default file names depend on mode: description.json, categorization.json, moderation.json, custom.json, or search.json. |
Request Example
Produces Video Intelligence output JSON
{
"query": {
"source": "https://your-storage.com/source.mov",
"encoder_version": "2",
"format": [
{
"output": "video_intelligence",
"mode": "description",
"result_name": "description.json"
},
{
"output": "video_intelligence",
"mode": "categorization",
"result_name": "categorization.json",
"categories": [
"Places",
"Nature",
"Vlogs"
]
},
{
"output": "video_intelligence",
"mode": "moderation",
"result_name": "moderation.json",
"violation_reasons": [
"Dangerous Acts",
"Graphic Violence",
"Symbolism of Hate"
]
},
{
"output": "video_intelligence",
"mode": "custom",
"result_name": "custom.json",
"prompt": "prompt"
},
{
"output": "video_intelligence",
"mode": "search",
"result_name": "search.json",
"prompt": "prompt",
"media_prompt": "image_url",
"search_options": [
"visual",
"audio",
"transcription"
],
"search_rank_threshold": 100
}
]
}
}Copy this payload and use it with your API key to test Video Intelligence on your own video.
View Video Intelligence TutorialPricing
What It Costs
Usage-Based Pricing
Video Intelligence is available for all Qencode users. Pricing is based on the selected analysis mode and usage, so you only pay for the videos you analyze.
We love creating powerful solutions that are aligned with the needs of your business.
Please send us a message if you have a question or Schedule a call for a demo to discuss your integration.
Let's talk
Contact us with any questions. We'd love to help.
Los Angeles, CA - (HQ)
San Francisco, CA
New York, NY