Exploring the synergies of hybrid convolutional neural network and Vision Transformer architectures for computer vision: A survey

Yunusa Haruna, Shiyin Qin, Abdulrahman Hamman Adama Chukkol, Abdulganiyu Abdu Yusuf, Isah Bello, Adamu Lawan

Abstract: Highlights•Overview of CNN and ViT evolution, highlighting strengths in CV tasks.•Taxonomy of hybrid CNN-ViT models: parallel, serial, early and late fusion, MHSA.•Comparative analysis and application of hybrid designs for CV tasks.•Challenges: design complexity, computational cost, and future research.•Research conclusion for researchers and developers.

External IDs:doi:10.1016/j.engappai.2025.110057