Performance-Efficiency Trade-off for Fashion Image Retrieval

2025-09-29

AI Summary

To address the high computational cost and poor scalability of secondhand clothing image retrieval systems over large-scale databases, this paper proposes a selective representation framework. The core innovations include: (i) a neighborhood homogeneity consistency scoring mechanism for efficient outlier removal, and (ii) a joint clustering–core-set selection strategy to identify the most representative image embedding vectors. Crucially, the method operates without requiring additional annotations, significantly compressing the vector database size. Experiments on three public benchmarks demonstrate that, when the database is reduced to 10% of its original size, retrieval accuracy remains near-optimal—mean Average Precision (mAP) degrades by less than 1.2%—while inference speed improves by approximately 9×. This work provides a scalable, high-accuracy, low-overhead solution for fine-grained cross-domain image retrieval.

Technology Category

• Computer Vision: Image and Video Retrieval
• Machine Learning: Dimensionality Reduction/Feature Selection
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs

Abstract

The fashion industry has been identified as a major contributor to waste and emissions, leading to an increased interest in promoting the second-hand market. Machine learning methods play an important role in facilitating the creation and expansion of second-hand marketplaces by enabling the large-scale valuation of used garments. We contribute to this line of work by addressing the scalability of second-hand image retrieval from databases. By introducing a selective representation framework, we can shrink databases to 10% of their original size without sacrificing retrieval accuracy. We first explore clustering and coreset selection methods to identify representative samples that capture the key features of each garment and its internal variability. Then, we introduce an efficient outlier removal method, based on a neighbour-homogeneity consistency score measure, that filters out uncharacteristic samples prior to selection. We evaluate our approach on three public datasets: DeepFashion Attribute, DeepFashion Con2Shop, and DeepFashion2. The results demonstrate a clear performance-efficiency trade-off by strategically pruning and selecting representative vectors of images. The retrieval system maintains near-optimal accuracy, while greatly reducing computational costs by reducing the images added to the vector database. Furthermore, applying our outlier removal method to clustering techniques yields even higher retrieval performance by removing non-discriminative samples before the selection.

Problem

• Addressing scalability of second-hand fashion image retrieval
• Shrinking databases without sacrificing retrieval accuracy
• Reducing computational costs while maintaining near-optimal performance

Innovation

• Selective representation framework shrinks database size
• Clustering and coreset selection identify representative garment samples
• Outlier removal method filters uncharacteristic samples before selection