# Image-Text Knowledge Modeling for Unsupervised Multi-Scenario Person Re-Identification

arXiv: https://arxiv.org/abs/2601.11243 · HTML: https://mlsanity.com/paper/2601.11243/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 23 references that are in the map.
- First posted on arXiv 16 Jan 2026.

## Builds on (most cited first, 11 of 11)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Deep Learning for Person Re-identification: A Survey and Outlook](https://mlsanity.com/paper/2001.04193/) (2020)
- [CLIP-ReID: Exploiting Vision-Language Model for Image Re-Identification without Concrete Text Labels](https://mlsanity.com/paper/2211.13977/) (2022)
- [Cluster Contrast for Unsupervised Person Re-Identification](https://mlsanity.com/paper/2103.11568/) (2021)
- [Long-Term Cloth-Changing Person Re-identification](https://mlsanity.com/paper/2005.12633/) (2020)
- [Camera-aware Proxies for Unsupervised Person Re-Identification](https://mlsanity.com/paper/2012.10674/) (2020)
- [Intra-Inter Camera Similarity for Unsupervised Person Re-Identification](https://mlsanity.com/paper/2103.11658/) (2021)
- [RLE: A Unified Perspective of Data Augmentation for Cross-Spectral Re-identification](https://mlsanity.com/paper/2411.01225/) (2024)
- [Revisiting Few-Shot Learning from a Causal Perspective](https://mlsanity.com/paper/2209.13816/) (2022)
- [SiCL: Silhouette-Driven Contrastive Learning for Unsupervised Person Re-Identification with Clothes Change](https://mlsanity.com/paper/2305.13600/) (2023)

## Abstract (arXiv)

We propose unsupervised multi-scenario (UMS) person re-identification (ReID) as a new task that expands ReID across diverse scenarios (cross-resolution, clothing change, etc.) within a single coherent framework. To tackle UMS-ReID, we introduce image-text knowledge modeling (ITKM) -- a three-stage framework that effectively exploits the representational power of vision-language models. We start with a pre-trained CLIP model with an image encoder and a text encoder. In Stage I, we introduce a scenario embedding in the image encoder and fine-tune the encoder to adaptively leverage knowledge from multiple scenarios. In Stage II, we optimize a set of learned text embeddings to associate with pseudo-labels from Stage I and introduce a multi-scenario separation loss to increase the divergence between inter-scenario text representations. In Stage III, we first introduce cluster-level and instance-level heterogeneous matching modules to obtain reliable heterogeneous positive pairs (e.g., a visible image and an infrared image of the same person) within each scenario. Next, we propose a dynamic text representation update strategy to maintain consistency between text and image supervision signals. Experimental results across multiple scenarios demonstrate the superiority and generalizability of ITKM; it not only outperforms existing scenario-specific methods but also enhances overall performance by integrating knowledge from multiple scenarios.