Efficient virtual data search for annotation‐free vehicle reidentification

Abstract

Vehicle reidentification (re‐ID) is the task of retrieving the same vehicle across nonoverlapping cameras, which has made significant progress with the help of abundant manually annotated real images. To avoid the time‐consuming and tedious labeling of real images, virtual data sets with large‐scale synthetic images have recently been constructed to perform annotation‐free model training. However, current methods fail to exploit the potential of virtual data search, that is, searching valuable and representative virtual subdata set for efficient training. This paper presents a novel data sampling strategy from both semantic and feature levels to perform an effective data search. The semantic level determines the sample number of each vehicle identity via the consistency constraint of attribute distribution for source domain and target domain; while the feature level searches valuable and representative samples of each vehicle identity. To our knowledge, we are among the first attempts to search effective virtual data to perform annotation‐free vehicle re‐ID. Extensive cross‐domain experiments from virtual vehicle re‐ID data sets to real vehicle re‐ID data sets show that our data sampling strategy can significantly reduce the training data volume and even boost the re‐ID performance.

Contributing Dimension Structure of Deep Feature for Coreset Selection

“…Coreset selection is a long-standing learning problem that aims to select a subset of the most informative training samples for data-efficient learning (Das et al 2021;Wan et al 2023a). Early coreset selection methods were designed to accelerate the learning and clustering of machine algorithms, such as k-means and k-medians (Har-Peled and Kushal 2007), support vector machines (Tsang et al 2005), (1,1)…”

“…Diversity Measurement Informative and easily accessible deep features are often adopted in selection methods (Guo, Zhao, and Bai 2022;Margatina et al 2021;Wan et al 2023bWan et al , 2022, which typically use the overall similarity metrics to calculate the similarity between deep features to measure the importance of the data, such as L1-norm, L2norm and cosine distance metric. Figure 2 illustrates their characteristics.…”

Research on Network Information Security Risk Assessment Based on Artificial Intelligence

“…Diversity Measurement Informative and easily accessible deep features are often adopted in selection methods (Guo, Zhao, and Bai 2022;Margatina et al 2021;Wan et al 2023bWan et al , 2022, which typically use the overall similarity metrics to calculate the similarity between deep features to measure the importance of the data, such as L1-norm, L2norm and cosine distance metric. Figure 2 illustrates their characteristics.…”