Features · Visual Search

Describe what you need,
find the exact frame

Type what you're looking for in plain language. Search video by scene, color, or description and FrameQuery matches your query against every frame in your library using AI visual embeddings and semantic transcript analysis.

How It Works

Two AI models, one search bar

Visual search and semantic transcript search run simultaneously when you type a query. Results are ranked by combined relevance so the best matches surface first.

sunset over water
All indexes2

3 results · 12ms

94%01:14:22

A001_C003.mov

Visual
87%00:42:08

B_ROLL_LAKE.mp4

Transcript
72%00:03:51

INTERVIEW_EXT.mov

Object

Find Similar

Pick a scene, find every scene like it

Click “Find Similar” on any scene card to search for visually similar scenes across your library. Uses cosine distance between SigLIP embeddings to rank results by visual similarity.

Find Similar Scenes
This VideoAll Videos

Source Scene

26s - 28s
aerialaerial

An aerial view shows a vast, dry grassland dotted with scattered trees and rounded hills under a bright, cloudy sky.

HillTreeGrasslandSky
9m 02s - 9m 18s94%

A vast golden field of dry grass with a dirt road cutting through, leading towards distant rolling hills.

DescriptionShotColor
18m 12s - 19m 32s87%

Savannah landscape with scattered acacia trees stretching to the horizon under golden afternoon light.

DescriptionAngle
3m 45s - 4m 12s79%

A large herd of animals moving in a winding line across dry grassland with scattered trees.

DescriptionShotColor
31m 08s - 31m 44s68%

Wide aerial shot of river winding through arid plains at dusk with long shadows.

ShotAngle

Search Features

Search by color, scene, or object

Search video by color to find scenes that match a specific palette. Search by scene description to locate the right mood, setting, or composition. Object detection tags every scene with what appears on screen, so you can search for “laptop” or “car” directly. Find visually similar shots by clicking any scene card. Everything runs locally and offline with no cloud dependency.

About the visual search models

Visual search models are optional and download on demand. SigLIP handles image-text matching, MiniLM handles semantic text similarity. Both are ONNX format and run with CUDA or Metal acceleration when available, with CPU fallback. Your search index works without them (keyword search only) until you choose to enable visual search.