Provided is a training method for updating a deep learning-based object detector of an autonomous vehicle so as to adapt to a driving environment, the method comprising (a) (i) inputting a training image corresponding to the driving environment into an environment-specific object detector so as to cause the k-th environment-specific object detector (i-1) to generate an environment-specific feature map in a convolution manner, (i-2) generating an environment-specific pooled feature map using ROI pooling, (i-3) applying a full connection operation to the kth environment-specific pooled feature map to generate environment-specific object detection information, (ii) inputting the environment-specific feature map into a specific environment-specific sorting network, (ii-1) performing a deconvolution operation through an environment-specific deconvolution layer to generate an environment-specific segmentation map, and (ii-2) outputting the environment-specific segmentation map to the k-th environment-specific pooled feature map to the k-th environment-specific pooled feature map. (ii-2) generating an environment-specific ranking score by an environment-specific discriminator, (b) training an environment-specific object detector, training an environment-specific deconvolution layer, training an environment-specific convolution layer, training an environment-specific discriminator.

    提供用于更新自动驾驶汽车的基于深度学习的对象检测器使得适应行驶环境的训练方法,该方法包括(a)(i)将对应于驾驶环境的训练图像输入到环境特定对象检测器中,以使所述第k环境特定对象检测器(i‑1)卷积生成环境特定特征图,(i‑2)利用ROI池化生成环境特定池化特征图,(i‑3)对所述第k环境特定池化特征图适用全连接运算生成环境特定对象检测信息,(ii)将环境特定特征图输入到特定环境特定排序网络,(ii‑1)通过环境特定去卷积层进行去卷积运算生成环境特定分割图,(ii‑2)通过环境特定鉴别器生成环境特定排序分数,(b)训练环境特定对象检测器,训练环境特定去卷积层环境特定卷积层,训练环境特定鉴别器。


    Access

    Download


    Export, share and cite



    Title :

    TRAINING METHOD AND LEARNING APPARATUS FOR UPDATING DEEP LEARNING-BASED OBJECT DETECTOR OF AUTONOMOUS DRIVING AUTOMOBILE TO ADAPTED TO DRIVING ENVIRONMENT,


    Additional title:

    用于更新自动驾驶汽车的基于深度学习的对象检测器使得适应行驶环境的训练方法和学习设备、以及利用该方法的更新方法和更新设备


    Contributors:
    RYU WOOJU (author) / JE HONGMO (author) / KANG FENGNAN (author) / KIM YONGJOONG (author)

    Publication date :

    2022-12-23


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion / G05D SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES , Systeme zum Steuern oder Regeln nichtelektrischer veränderlicher Größen / G06V