03 January 2024 0 5K Report

Implicit neural representation (INR, sometimes also referred to as coordinate-based representation) aims to estimate implicit function that represents a signal (e.g. image, sound) continuously. In the most of works based on INR, they adopt MLP rather than CNN as the "implicit function". But why? I‘m really confused about it. Hope someone can help me to understand it. Thanks in advance.

More Yaohang Wu's questions See All
Similar questions and discussions