Method and device for generalizing an image classification model for a vehicle driver assistance system

Abstract

The present invention relates to a method and device for generalizing an image classification model for a vehicle driver assistance system. To this end, a test image is augmented to obtain a first highly augmented image (101) and a second slightly augmented image (102). Contour images (103, 104) are generated for the first highly augmented image (101) and the second slightly augmented image (102). Vector representations (111, 112, 121, 122) of the features of each of the augmented images (101, 102) and of each of the contour images (103, 104) are determined by a first encoder (11) and a second encoder (12). The parameters of the classification model are updated by minimizing at least one loss function, including contrastive loss, from the vector representations (111, 112, 121, 122). Figure for the abstract: Figure 1

Claims

traffic of said vehicle acquired by a camera mounted in said vehicle; - classification of said image data by said classification model mounted in said vehicle; and - control of the driver assistance system according to a result of said classification.

8. A computer program comprising instructions for carrying out the method according to any one of the preceding claims, when such instructions are executed by a processor.

9. A computer-readable recording medium on which is recorded a computer program comprising instructions for carrying out the steps of the process according to any one of claims 1 to 7

10. 1 d / . Device (2) for generalizing an image classification model for a vehicle driver assistance system, said device (2) comprising a memory (21) associated with at least one processor (20) configured for the implementation of at least one step of the method according to any one of claims 1 to 7. [Fig. 1] 1 neither 121 [Fig. 2] 2

Images