Deploying state-of-the-art object detectors on resource-limited devices presents significant challenges. Knowledge distillation is an efficient and streamlined lightweight technique to improve the accuracy of compact detectors. However, its effectiveness is limited by the redundancy of different types of semantics on the feature map and the closure of same level’s feature distillation. To alleviate this problem, we propose Decoupled and Interactive Distillation, an effective and versatile method to improve knowledge distillation in some complex object detection tasks. The method has two key components. A knowledge decoupled module captures category awareness and localization awareness features. A multi-level feature interaction distillation can aggregate feature distillations from shallow to deep levels, facilitating the collaboration between feature transfers at different levels. The relevant experiments in traffic-related, 3D, rotated object detection have verified the effectiveness of the proposed method, particularly in challenging scenes.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Empowering Object Detection: Unleashing the Potential of Decoupled and Interactive Distillation


    Contributors:
    Qian, Fulan (author) / Hong, Jiacheng (author) / Yan, Huanqian (author) / Chen, Hai (author) / Zhang, Chonghao (author) / Su, Hang (author) / Zhao, Shu (author)


    Publication date :

    2025-01-01


    Size :

    5822906 byte




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English



    Scale Decoupled Distillation

    Luo, Shicai Wei Chunbo Luo Yang | ArXiv | 2024

    Free access

    Decoupled Knowledge Distillation

    Zhao, Borui / Cui, Quan / Song, Renjie et al. | ArXiv | 2022

    Free access

    UNLEASHING AIR FREIGHT'S POTENTIAL

    FULLER, E. W. / STOESSEL, R. F. | SAE Technical Papers | 1958


    Unleashing air freight's potential

    Fuller, E.W. / Stoessel, R.F. | Engineering Index Backfile | 1958