Search papers, labs, and topics across Lattice.
This survey comprehensively reviews the advancements in multi-label image classification (MLIC) using deep learning, categorizing approaches into six distinct groups based on their methodologies. It highlights the significant improvements in robustness, scalability, and generalization of MLIC models due to deep learning techniques, which are crucial for applications like autonomous driving and disease diagnosis. The paper also discusses the challenges and future directions for MLIC, providing a structured framework that can guide further research and innovation in the field.
A systematic taxonomy of multi-label image classification methods reveals critical insights and challenges that could redefine future research trajectories in computer vision.
Multi-label image classification (MLIC), a fundamental task in computer vision, focuses on identifying multiple objects or concepts within an image, underpinning numerous read-world applications, such as autonomous driving, disease diagnosis, recommendation system, and mobile service robot. Over the past decade, deep learning paradigms based on convolutional neural networks, recurrent neural networks, and Transformers have significantly advanced this field, owing to their powerful capability in visual representation and relationship modeling. These advances have markedly improved the robustness, scalability, and generalization ability of MLIC models across diverse datasets and application domains. In this survey, we provide a comprehensive review of the deep learning-based literature on MLIC. Concretely, we first revisit the background, including problem definition, datasets, backbones and evaluation metrics. Next, we develop a plausible taxonomy for the deep learning-based MLIC approaches, organizing them into six groups: region-oriented methods, label-oriented methods, architecture-oriented methods, representation-oriented methods, learning-oriented methods, and data-oriented methods. Finally, we provide an insightful exposition of the underlying learning game in MLIC and its implications for other vision domains, and we empirically summarize the key challenges and research directions in MLIC while outlining promising avenues for future development. We believe this survey offers the research community a holistic and systematic perspective on MLIC, thereby facilitating subsequent exploration and innovation in this field and beyond.
