Skip to main content

​
Video Quality Detectors

​
video_specs

Validates video technical specifications.
​
config.min_width
integer
Minimum video width in pixels
​
config.max_width
integer
Maximum video width in pixels
​
config.min_height
integer
Minimum video height in pixels
​
config.max_height
integer
Maximum video height in pixels
​
config.allowed_aspect_ratios
array
List of allowed aspect ratios (e.g., [“16:9”, “9:16”, “1:1”])
​
config.min_duration_sec
number
Minimum video duration in seconds
​
config.max_duration_sec
number
Maximum video duration in seconds
​
config.min_fps
number
Minimum frames per second
​
config.max_fps
number
Maximum frames per second
Example: 
{
  "detector_name": "video_specs",
  "config": {
    "min_width": 1920,
    "min_height": 1080,
    "min_duration_sec": 15,
    "max_duration_sec": 60,
    "min_fps": 24
  }
}

​
video_quality

Checks video quality metrics.
​
config.expected_aspect_ratios
array
Expected aspect ratios
​
config.check_aspect_ratio
boolean
Whether to check aspect ratio consistency
​
config.min_width
integer
Minimum width in pixels
​
config.min_height
integer
Minimum height in pixels

​
Image Detectors

​
image_specs

Validates image specifications.
​
config.allowed_aspect_ratios
array
List of allowed aspect ratios
​
config.min_width
integer
Minimum image width in pixels
​
config.min_height
integer
Minimum image height in pixels
​
config.min_megapixels
number
Minimum image size in megapixels

​
Content Matching Detectors

​
video_matching

Matches video against reference content. Requires preprocessing.
​
config.reference_videos
array
required
Array of GCS URIs of reference videos (gs://bucket/path/to/video.mp4)Allowed extensions: mp4, mov, avi, mkv, webm
Example: 
{
  "detector_name": "video_matching",
  "config": {
    "reference_videos": [
      "gs://judge-artifacts/org_123/reference_uploads/abc/video1.mp4",
      "gs://judge-artifacts/org_123/reference_uploads/def/video2.mp4"
    ]
  }
}

​
audio_matching

Matches audio against reference content. Requires preprocessing.
​
config.reference_audio
string
required
GCS URI of reference audio (gs://bucket/path/to/audio.mp3)Allowed extensions: mp3, wav, aac, flac

​
splitscreen_video_matching

Matches split-screen video content. Requires preprocessing.
​
config.source_video_references
array
Array of GCS URIs of source video referencesAllowed extensions: mp4, mov, avi, mkv, webm
​
config.source_video_description
string
Text description of source video
​
config.bottom_content_references
array
Array of GCS URIs of bottom content referencesAllowed extensions: mp4, mov, avi, mkv, webm
​
config.top_content_description
string
Text description of top content
Note: At least one of source_video_references or bottom_content_references must be provided.

​
video_matching_description

Matches video against text description.
​
config.content_description
string
required
Text description of expected content

​
Text Analysis Detectors

​
language

Validates content language.
​
config.accepted_languages
array
required
List of accepted language codes (e.g., [“en”, “es”, “fr”])

​
transcript_policy

Checks transcript against policy guidelines.
​
config.talking_points
array
List of required talking points
​
config.language_to_avoid
array
List of prohibited words or phrases
​
config.min_talking_points
integer
Minimum number of talking points required
​
config.max_language_to_avoid
integer
Maximum allowed violations

​
transcript_product

Detects product mentions.
​
config.brand_name
string
Brand name to detect
​
config.product_name
string
Product name to detect

​
sentiment

Analyzes emotional tone.
​
config.desired_sentiment
array
List of desired sentiments (e.g., [“positive”, “enthusiastic”])
​
config.undesired_sentiment
array
List of undesired sentiments (e.g., [“negative”, “angry”])
​
config.prohibited_content_types
array
List of prohibited content types

​
Visual Analysis Detectors

​
overlays

Detects text overlays and graphics.
​
config.disallowed_overlays
array
List of overlay categories that should NOT appear (e.g., [“text_overlay”, “sticker”, “watermark”, “graphic”])
​
config.required_overlays
array
List of specific overlay descriptions that MUST appear (e.g., [“text showing #ad”, “heart emoji”])

​
wardrobe

Analyzes clothing and appearance.
​
config.wardrobe_rules
array
List of wardrobe requirements
​
config.beauty_rules
array
List of beauty/appearance requirements
​
config.max_violations
integer
Maximum allowed violations

​
product_application

Detects product usage demonstrations.
​
config.require_application
boolean
Whether product application demonstration is required

​
subtitles

Detects subtitle presence.
​
config.require_subtitles
boolean
Whether subtitles are required

​
image_ip_infringement

Checks for potential IP violations.
​
config.allowed_terms
array
List of allowed trademarked terms

​
Audio Analysis Detectors

​
pronunciation

Checks brand/product name pronunciation.
​
config.brand_name
string
Brand name to check
​
config.product_name
string
Product name to check
​
config.brand_cue
string
Phonetic cue for brand name
​
config.product_cue
string
Phonetic cue for product name
​
config.brand_audio_url
string
GCS URI of reference audio for brand nameAllowed extensions: mp3, wav, aac, flac, m4a
​
config.product_audio_url
string
GCS URI of reference audio for product nameAllowed extensions: mp3, wav, aac, flac, m4a

​
voice_over

Detects voice-over narration.
​
config.fail_if_voice_over
boolean
Whether to fail if voice-over is detected

​
Metadata Detectors

​
title_and_description

Analyzes hashtags and social media metadata.
​
config.suggested_hashtags
array
List of suggested hashtags
​
config.required_hashtags
array
List of required hashtags
​
config.required_social_tags
array
List of required social media tags
​
config.required_social_tags_mode
string
Mode for social tag requirements (“all” or “any”)
​
config.required_text
array
List of text that must appear in title or description
​
config.required_text_mode
string
Mode for required text (“all” or “any”)
​
config.topic_keywords
array
List of topic keywords to detect
​
config.enable_topic_analysis
boolean
Whether to enable topic analysis

​
transition_detection

Detects scene transitions and cuts.
​
config.max_transitions
integer
Maximum allowed scene transitions

​
Detectors Without Configuration

The following detectors work with default settings and don’t require configuration:
  • audio_noise - Detects background noise and audio quality issues
  • background - Analyzes background elements
  • content_quality - Evaluates overall content quality
  • framing - Evaluates shot composition and framing
  • image_sharpness - Detects blurry or out-of-focus images
  • lighting - Analyzes lighting quality
These can be included with an empty config object: 
{
  "detector_name": "audio_noise",
  "config": {}
}