RSTPReid

The 'RSTPReid' dataset/benchmark is a large-scale person image database used to evaluate text-based person retrieval methods, particularly focusing on the challenges of color over-reliance in cross-modal data matching.

Papers using RSTPReid (10)

Harnessing Weak Pair Uncertainty for Text-based Person Search2026GEA: Generation-Enhanced Alignment for Text-to-Image Person Retrieval2025CAMeL: Cross-modality Adaptive Meta-Learning for Text-based Person Retrieval2025Multi-modal Reference Learning for Fine-grained Text-to-Image Retrieval2025See Finer, See More: Implicit Modality Alignment for Text-based Person Retrieval2022 · 9 citesSCMM: Calibrating Cross-modal Representations for Text-Based Person Search2023 · 6 citesRaSa: Relation and Sensitivity Aware Representation Learning for Text-based Person Search2023 · 6 citesTowards Unified Text-based Person Retrieval: A Large-scale Multi-Attribute and Language Search Benchmark2023 · 6 citesText-based Person Search without Parallel Image-Text Data2023 · 1 citesCAIBC: Capturing All-round Information Beyond Color for Text-based Person Retrieval2022