Boosting Weak Positives For Text Based Person Search

Abstract

Large vision-language models have revolutionized cross-modal object retrieval, but text-based person search (TBPS) remains a challenging task due to limited data and fine-grained nature of the task. Existing methods primarily focus on aligning image-text pairs into a common representation space, often disregarding the fact that real world positive image-text pairs share a varied degree of similarity in between them. This leads models to prioritize easy pairs, and in some recent approaches, challenging samples are discarded as noise during training. In this work, we introduce a boosting technique that dynamically identifies and emphasizes these challenging samples during training. Our approach is motivated from classical boosting technique and dynamically updates the weights of the weak positives, wherein, the rank-1 match does not share the identity of the query. The weight allows these misranked pairs to contribute more towards the loss and the network has to pay more attention toward

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keymodi2025boosting

Related papers

• Text-based Person Search With Limited Data (2021)15.69
• Contrastive Transformer Learning With Proximity Data Generation For Text-based Person Search (2023)11.88
• Semi-supervised Text-based Person Search (2024)3.58
• Beat: Bi-directional One-to-many Embedding Alignment For Text-based Person Retrieval (2024)10.85
• Enhancing Visual Representation For Text-based Person Searching (2024)1.69
• TIPCB: A Simple But Effective Part-based Convolutional Baseline For Text-based Person Search (2021)20.24
• Deep Boosting Learning: A Brand-new Cooperative Approach For Image-text Matching (2024)9.73
• Multi-path Exploration And Feedback Adjustment For Text-to-image Person Retrieval (2024)0.00