An Analysis of Multi-Task Architectures for the Hierarchic Multi-Label Problem of Vehicle Model and Make Classification

Researchers analyzed multi-task learning architectures for hierarchical classification of vehicle makes and models, testing CNN and Transformer models on StanfordCars and CompCars datasets. The study found that multi-task approaches improved performance for CNNs in almost all scenarios and yielded significant improvements for both model types on the CompCars dataset.

• →Multi-task learning architectures can effectively leverage hierarchical structure in classification problems.
• →CNN models showed performance improvements in almost all tested scenarios using multi-task approaches.
• →Both parallel and cascaded multi-task architectures were evaluated across different deep learning models.
• →The CompCars dataset showed significant improvements for both CNN and Transformer models.
• →Research demonstrates that exploiting hierarchical information structure can enhance deep learning model performance.