SMMix: Self-Motivated Image Mixing for Vision Transformers

doi:doi:10.57702/clz8l3fh

SMMix: Self-Motivated Image Mixing for Vision Transformers

CutMix is a vital augmentation strategy that determines the performance and generalization ability of vision transformers (ViTs). However, the inconsistency between the mixed images and the corresponding labels harms its effectiveness. Existing CutMix variants tackle this problem by generating more consistent mixed images or more precise mixed labels, but inevitably introduce heavy training overhead or require extra information, undermining ease of use.

BibTex:

@dataset{Mengzhao_Chen_and_Mingbao_Lin_and_Zhihang_Lin_and_Yuxin_Zhang_and_Fei_Chao_and_Rongrong_Ji_2024,
    abstract = {CutMix is a vital augmentation strategy that determines the performance and generalization ability of vision transformers (ViTs). However, the inconsistency between the mixed images and the corresponding labels harms its effectiveness. Existing CutMix variants tackle this problem by generating more consistent mixed images or more precise mixed labels, but inevitably introduce heavy training overhead or require extra information, undermining ease of use.},
    author = {Mengzhao Chen and Mingbao Lin and Zhihang Lin and Yuxin Zhang and Fei Chao and Rongrong Ji},
    doi = {10.57702/clz8l3fh},
    institution = {No Organization},
    keyword = {'CutMix', 'SMMix', 'image mixing', 'vision transformers'},
    month = {dec},
    publisher = {TIB},
    title = {SMMix: Self-Motivated Image Mixing for Vision Transformers},
    url = {https://service.tib.eu/ldmservice/dataset/smmix--self-motivated-image-mixing-for-vision-transformers},
    year = {2024}
}