# Sense-X/Co-DETR

[ICCV 2023] DETRs with Collaborative Hybrid Assignments Training

Repository: https://github.com/Sense-X/Co-DETR
Canonical: https://ross.abutalabs.com/products/co-detr
Language: Python
License: MIT
License Family: permissive
Last push: 2024-12-29T10:11:47+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 98
- inputs: {"age_days": 1383, "days_push": 612, "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 1357, forks 180 (observed 2026-08-28T04:04:29.472510+00:00)

## What it is
Co-DETR is a PyTorch implementation of DETRs with Collaborative Hybrid Assignments Training, an ICCV 2023 object detection and instance segmentation framework. It provides pretrained models and configs that achieve state-of-the-art results on COCO and LVIS benchmarks.

## Use cases
- train state-of-the-art object detection models on COCO
- run instance segmentation with DETR-based transformers
- fine-tune a ViT-L detector on custom datasets
- get pretrained detection checkpoints from Hugging Face
- benchmark DETR variants on LVIS
- use a strong vision backbone for multimodal models

## When to choose
- you need top-accuracy object detection or instance segmentation with transformer detectors
- you want reproducible configs and pretrained weights for COCO/LVIS research
- you need a strong vision encoder for downstream multimodal tasks

## When to avoid
- you need lightweight real-time detection on edge devices
- you want a plug-and-play production API rather than a research codebase
- you work outside the PyTorch/MMDetection ecosystem

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, deep-learning, computer-vision, image-processing
- domain: computer-vision, deep-learning, machine-learning, artificial-intelligence
- platform: python
- tags: object-detection, detr, instance-segmentation, transformer, iccv-2023, coco, lvis, research, linux, gpu

## Member repositories
- Sense-X/Co-DETR (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:29.472510+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 2026-08-30T04:41:48.787915+00:00, confidence not recorded.
  - readme: https://github.com/Sense-X/Co-DETR (fetched 2026-08-28T04:04:29.472510+00:00, sha 39854c03ea84)
- Data as of 2026-08-30T08:39:29.467469+00:00.