Scene-centric Vs. Object-centric Image-text Cross-modal Retrieval: A Reproducibility Study

Abstract

Most approaches to cross-modal retrieval (CMR) focus either on object-centric datasets, meaning that each document depicts or describes a single object, or on scene-centric datasets, meaning that each image depicts or describes a complex scene that involves multiple objects and relations between them. We posit that a robust CMR model should generalize well across both dataset types. Despite recent advances in CMR, the reproducibility of the results and their generalizability across different dataset types has not been studied before. We address this gap and focus on the reproducibility of the state-of-the-art CMR results when evaluated on object-centric and scene-centric datasets. We select two state-of-the-art CMR models with different architectures: (i) CLIP; and (ii) X-VLM. Additionally, we select two scene-centric datasets, and three object-centric datasets, and determine the relative performance of the selected models on these datasets. We focus on reproducibility, replicability,

Stats

• citations4
• S2 citations—
• github stars0
• HF likes0
• heat score5.24
• arxiv keyhendriksen2023scene

Related papers

• Stacmr: Scene-text Aware Cross-modal Retrieval (2020)10.48
• A Comprehensive Empirical Study Of Vision-language Pre-trained Model For Supervised Cross-modal Retrieval (2022)0.00
• Rethinking Benchmarks For Cross-modal Image-text Retrieval (2023)13.11
• Evaluating Perspectival Biases In Cross-modal Retrieval (2025)0.00
• Caption-matching: A Multimodal Approach For Cross-domain Image Retrieval (2024)0.00
• Revisiting Cross Modal Retrieval (2018)0.00
• Deep Reversible Consistency Learning For Cross-modal Retrieval (2025)7.81
• Where Does The Performance Improvement Come From? -- A Reproducibility Concern About Image-text Retrieval (2022)3.36