Vision encoders should be image size agnostic and task driven
It addresses efficiency issues in vision encoders for AI/ML applications, but is incremental as it presents a conceptual argument with initial steps rather than a full solution.
This position paper argues that vision encoders should be image size agnostic and task-driven to improve efficiency, inspired by biological vision, and provides a proof-of-concept for image classification to demonstrate feasibility.
This position paper argues that the next generation of vision encoders should be image size agnostic and task driven. The source of our inspiration is biological. Not a structural aspect of biological vision, but a behavioral trait -- efficiency. We focus on a couple of ways in which vision in nature is efficient, but modern vision encoders not. We -- humans and animals -- deal with vast quantities of visual data, and need to be smart where we focus our limited energy -- it depends on the task. It is our belief that vision encoders should be dynamic and the computational complexity should depend on the task at hand rather than the size of the image. We, also, provide concrete first steps towards our vision -- a proof-of-concept solution for image classification. Despite classification being not very representative for what we are trying to achieve, it shows that our approach is feasible and promising.