# microsoft/VideoX

VideoX: a collection of video cross-modal models

Repository: https://github.com/microsoft/VideoX
Canonical: https://ross.abutalabs.com/products/videox
Language: Python
License: NOASSERTION
License Family: other
Last push: 2024-06-03T02:11:26+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2477, "days_push": 822, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1071, forks 165 (observed 2026-08-28T04:03:28.256314+00:00)

## What it is
VideoX is a collection of Microsoft's video cross-modal understanding models, including X-CLIP for video-language recognition, 2D-TAN and MS-2D-TAN for natural language moment localization, and SeqTrack for visual object tracking. It provides research code and pretrained models for video-text tasks.

## Use cases
- retrieve video moments using natural language queries
- classify videos with a language-image pretrained model
- track objects in video with a transformer model
- adapt CLIP to video recognition
- find timestamps in a video matching a text description

## When to choose
- you need pretrained video-text models like X-CLIP for research
- you want state-of-the-art moment localization or visual tracking baselines
- you are building on published CVPR/ECCV/AAAI video understanding methods

## When to avoid
- you need a production-ready video processing pipeline
- you want a maintained library with long-term support
- you need a permissive license for commercial use

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, computer-vision, video-processing, nlp
- domain: computer-vision, machine-learning
- platform: python
- tags: video-understanding, video-text-retrieval, moment-retrieval, visual-tracking, clip, research-code, transformer, video, natural-language-processing, linux, gpu

## Member repositories
- microsoft/VideoX (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:28.256314+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:53:50.968881+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/VideoX (fetched 2026-08-28T04:03:28.256314+00:00, sha 6d1dfc2765b1)
- Data as of 2026-08-30T08:39:29.467469+00:00.