# Towards Resolving Optimization Conflicts Between Image- and Text-Based Person Re-Identification

arXiv: https://arxiv.org/abs/2606.02242 · HTML: https://mlsanity.com/paper/2606.02242/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 30 references that are in the map.
- First posted on arXiv 1 Jun 2026.

## Builds on (most cited first, 23 of 23)

- [Person Transfer GAN to Bridge Domain Gap for Person Re-Identification](https://mlsanity.com/paper/1711.08565/) (2017)
- [TransReID: Transformer-based Object Re-Identification](https://mlsanity.com/paper/2102.04378/) (2021)
- [Joint Detection and Identification Feature Learning for Person Search](https://mlsanity.com/paper/1604.01850/) (2016)
- [CLIP-ReID: Exploiting Vision-Language Model for Image Re-Identification without Concrete Text Labels](https://mlsanity.com/paper/2211.13977/) (2022)
- [Person Search with Natural Language Description](https://mlsanity.com/paper/1702.05729/) (2017)
- [Cross-Modal Implicit Relation Reasoning and Aligning for Text-to-Image Person Retrieval](https://mlsanity.com/paper/2303.12501/) (2023)
- [Semantically Self-Aligned Network for Text-to-Image Part-aware Person Re-identification](https://mlsanity.com/paper/2107.12666/) (2021)
- [DSSL: Deep Surroundings-person Separation Learning for Text-based Person Retrieval](https://mlsanity.com/paper/2109.05534/) (2021)
- [Noisy-Correspondence Learning for Text-to-Image Person Re-identification](https://mlsanity.com/paper/2308.09911/) (2023)
- [RaSa: Relation and Sensitivity Aware Representation Learning for Text-based Person Search](https://mlsanity.com/paper/2305.13653/) (2023)
- [PLIP: Language-Image Pre-training for Person Representation Learning](https://mlsanity.com/paper/2305.08386/) (2023)
- [Instruct-ReID: A Multi-purpose Person Re-identification Task with Instructions](https://mlsanity.com/paper/2306.07520/) (2023)
- [PANDA: A Gigapixel-level Human-centric Video Dataset](https://mlsanity.com/paper/2003.04852/) (2020)
- [Large-Scale Pre-training for Person Re-identification with Noisy Labels](https://mlsanity.com/paper/2203.16533/) (2022)
- [MARS: Paying more attention to visual attributes for text-based person search](https://mlsanity.com/paper/2407.04287/) (2024)
- [Learnable Pillar-based Re-ranking for Image-Text Retrieval](https://mlsanity.com/paper/2304.12570/) (2023)
- [ReMix: Training Generalized Person Re-identification on a Mixture of Data](https://mlsanity.com/paper/2410.21938/) (2024)
- [A Culturally-Aware Benchmark for Person Re-Identification in Modest Attire](https://mlsanity.com/paper/2412.18874/) (2024)
- [ENTIRe-ID: An Extensive and Diverse Dataset for Person Re-Identification](https://mlsanity.com/paper/2405.20465/) (2024)
- [An Open-World, Diverse, Cross-Spatial-Temporal Benchmark for Dynamic Wild Person Re-Identification](https://mlsanity.com/paper/2403.15119/) (2024)
- [Hierarchical Prompt Learning for Image- and Text-Based Person Re-Identification](https://mlsanity.com/paper/2511.13575/) (2025)
- [DynaMix: Generalizable Person Re-identification via Dynamic Relabeling and Mixed Data Sampling](https://mlsanity.com/paper/2511.19067/) (2025)
- [ReText: Text Boosts Generalization in Image-Based Person Re-identification](https://mlsanity.com/paper/2602.05785/) (2026)

## Abstract (arXiv)

The joint optimization of image-based (I2I) and text-based (T2I) person re-identification (ReID) is hindered by modality discrepancies and conflicting training objectives, leading to suboptimal shared representations. While I2I ReID focuses on identity-level invariance across images of the same person, T2I ReID is driven by instance-specific textual descriptions tied to unique visual traits. This paper explores the fundamental difference between two ReID tasks and their optimization processes for effective training. Since I2I and T2I ReID are often studied separately, the loss functions optimized for one retrieval setting may negatively affect the representation quality required by the other. Motivated by these findings, we propose a decoupled two-stage training pipeline for learning a shared representation across image and text modalities. The pipeline is based on a single vision encoder that supports both I2I and T2I retrieval while avoiding cross-task interference during training. We provide extensive experiments across multiple configurations, varying domain mixing procedures, learning strategies, and task objectives. We observed that I2I ReID pre-training positively impacts the generalization ability to T2I data. Besides, we find that incorporating textual supervision during the vision encoder training stage enhances both I2I and T2I performance. We believe our insights provide a meaningful step toward unified ReID systems and cross-modal retrieval overall.