SemanticMIM: Marring Masked Image Modeling with Semantics Compression for General Visual Representation
This work addresses a domain-specific problem in computer vision by combining existing methods for incremental improvements in visual representation learning.
The paper tackles the problem of integrating masked image modeling (MIM) and contrastive learning (CL) for general visual representation by proposing SemanticMIM, which bridges compression and reconstruction phases to enhance performance and feature linear separability.
This paper represents a neat yet effective framework, named SemanticMIM, to integrate the advantages of masked image modeling (MIM) and contrastive learning (CL) for general visual representation. We conduct a thorough comparative analysis between CL and MIM, revealing that their complementary advantages fundamentally stem from two distinct phases, i.e., compression and reconstruction. Specifically, SemanticMIM leverages a proxy architecture that customizes interaction between image and mask tokens, bridging these two phases to achieve general visual representation with the property of abundant semantic and positional awareness. Through extensive qualitative and quantitative evaluations, we demonstrate that SemanticMIM effectively amalgamates the benefits of CL and MIM, leading to significant enhancement of performance and feature linear separability. SemanticMIM also offers notable interpretability through attention response visualization. Codes are available at https://github.com/yyk-wew/SemanticMIM.