# tinyvision/SOLIDER

A Semantic Controllable Self-Supervised Learning Framework to learn general human representations from massive unlabeled human images, which can benefit downstream human-centric tasks to the maximum extent

Repository: https://github.com/tinyvision/SOLIDER
Canonical: https://ross.abutalabs.com/products/solider
Language: Python
License: Apache-2.0
License Family: permissive
Topics: cvpr2023, self-supervised-learning, human-centric
Last push: 2023-07-21T08:22:12+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 97
- inputs: {"age_days": 1365, "days_push": 1139, "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 1504, forks 234 (observed 2026-08-28T04:04:54.779695+00:00)

## What it is
SOLIDER is a semantic-controllable self-supervised learning framework that learns general human representations from massive unlabeled human images. It provides pretrained backbones whose semantic/appearance information ratio can be tuned via a semantic controller to benefit downstream human-centric vision tasks.

## Use cases
- pretrain a backbone for person re-identification
- improve pedestrian attribute recognition models
- boost pedestrian detection on CityPersons
- fine-tune features for human pose estimation on COCO
- transfer learning for human parsing and semantic segmentation
- person search in surveillance imagery

## When to choose
- your downstream task is human-centric (re-ID, pose, pedestrian detection, person search)
- you lack labeled human images and want strong self-supervised pretrained features
- you need to control the semantic vs appearance trade-off in representations

## When to avoid
- your task involves general objects rather than humans
- you need a lightweight model for edge devices
- you require actively maintained software with frequent updates

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, computer-vision, image-processing
- domain: computer-vision, machine-learning, deep-learning
- platform: python
- tags: self-supervised-learning, human-centric, pretrained-model, cvpr2023, person-reidentification, pedestrian-detection, pose-estimation, linux, gpu

## Member repositories
- tinyvision/SOLIDER (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:54.779695+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 2026-08-30T04:32:46.658374+00:00, confidence not recorded.
  - readme: https://github.com/tinyvision/SOLIDER (fetched 2026-08-28T04:04:54.779695+00:00, sha 6b4aeb10d6fb)
- Data as of 2026-08-30T08:39:29.467469+00:00.