# FoundationVision/GLEE

[CVPR2024 Highlight]GLEE: General Object Foundation Model for Images and Videos at Scale

Repository: https://github.com/FoundationVision/GLEE
Canonical: https://ross.abutalabs.com/products/glee
Homepage: https://glee-vision.github.io/
Language: Python
License: MIT
License Family: permissive
Topics: foundation-model, object-detection, open-world, tracking, open-vocabulary-detection, open-vocabulary-segmentation, open-vocabulary-video-segmentation, referring-expression-comprehension, referring-expression-segmentation, video-instance-segmentation, video-object-segmentation, zero-shot-object-detection, referring-video-object-segmentation, interactive-segmentation, segment-anything
Last push: 2024-10-21T06:17:43+00:00

## Health v2 (maintenance only)
Score: 26/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 70
- inputs: {"age_days": 993, "days_push": 681, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1170, forks 77 (observed 2026-08-28T04:03:51.244651+00:00)

## What it is
GLEE is a general object foundation model for images and videos that unifies detection, segmentation, tracking, grounding, and open-world object identification in a single framework. It is a Python research codebase with pretrained checkpoints, trained on over ten million images with diverse supervision levels for strong zero-shot transfer.

## Use cases
- detect and segment arbitrary objects in images with open vocabulary
- track and segment objects in videos
- segment objects referred to by a text expression in images or videos
- perform zero-shot object detection on new datasets
- interactively segment objects given visual prompts
- identify and ground objects in the open world

## When to choose
- you need a single unified model for multiple object-centric vision tasks
- you want strong zero-shot generalization to unseen categories or datasets
- you need open-vocabulary detection, segmentation, or referring segmentation in one codebase
- you are doing research on object-level perception in images and videos

## When to avoid
- you need a lightweight production detector with minimal compute
- you only need a simple task-specific detector like plain YOLO inference
- you cannot run GPU inference or lack PyTorch experience
- you need fine-grained classification rather than object localization and segmentation

## Facets
- artifact type: library
- maturity: active
- function: computer-vision, image-processing, video-processing, machine-learning, deep-learning
- domain: computer-vision, image-processing, artificial-intelligence, machine-learning
- platform: python, cross-platform
- tags: object-detection, object-segmentation, open-vocabulary, zero-shot, multi-object-tracking, visual-grounding, foundation-model, cvpr-2024, referring-expression, video-segmentation, video, gpu

## Member repositories
- FoundationVision/GLEE (main) score 26

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:51.244651+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 2026-08-30T06:28:35.224810+00:00, confidence not recorded.
  - readme: https://github.com/FoundationVision/GLEE (fetched 2026-08-28T04:03:51.244651+00:00, sha 9e22d217c518)
  - homepage: https://glee-vision.github.io/ (fetched 2026-08-29T12:34:24.657197+00:00, sha 8b40568b5bf9)
- Data as of 2026-08-30T08:39:29.467469+00:00.