paper-with-me

홈 › Papers

Masked Frequency Modeling for Self-Supervised Visual Pre-Training

2022-06-15 · Jiahao Xie, Wei Li, Xiaohang Zhan, Ziwei Liu, Yew Soon Ong, Chen Change Loy

We present Masked Frequency Modeling (MFM), a unified frequency-domain-based approach for self-supervised pre-training of visual models. Instead of randomly inserting mask tokens to the input embeddings in the spatial domain, in this paper, we shift the perspective to the frequency domain. Specifically, MFM first masks out a portion of frequency components of the input image and then predicts the missing frequencies on the frequency spectrum. Our key insight is that predicting masked components in the frequency domain is more ideal to reveal underlying image patterns rather than predicting masked patches in the spatial domain, due to the heavy spatial redundancy. Our findings suggest that with the right configuration of mask-and-predict strategy, both the structural information within high-frequency components and the low-level statistics among low-frequency counterparts are useful in learning good representations. For the first time, MFM demonstrates that, for both ViT and CNN, a simple non-Siamese framework can learn meaningful representations even using none of the following: (i) extra data, (ii) extra model, (iii) mask token. Experimental results on image classification and semantic segmentation, as well as several robustness benchmarks show the competitive performance and advanced robustness of MFM compared with recent masked image modeling approaches. Furthermore, we also comprehensively investigate the effectiveness of classical image restoration tasks for representation learning from a unified frequency perspective and reveal their intriguing relations with our MFM approach.

📄 PDF Abstract BibTeX arXiv:2206.07706

Code (3)

jiahao000/mfm 공식 구현 pytorch
CoinCheung/MFM pytorch
weivision/correlational-image-modeling pytorch

Tasks

image-classificationImage ClassificationImage RestorationRepresentation LearningSemantic Segmentation

Similar Papers 제목 키워드 기반

The Devil is in the Frequency: Geminated Gestalt Autoencoder for Self-Supervised Visual Pre-Training

2022-04-18 · Hao liu, Xinghua Jiang, Xin Li, Antai Guo 외

The self-supervised Masked Image Modeling (MIM) schema, following "mask-and-reconstruct" pipeline of recovering contents from masked image, has recently captured the increasing interest in the multimedia community, owing…

Exploring the Coordination of Frequency and Attention in Masked Image Modeling

2022-11-28 · Jie Gui, Tuo Chen, Minjing Dong, Zhengqi Liu 외

Recently, masked image modeling (MIM), which learns visual representations by reconstructing the masked patches of an image, has dominated self-supervised learning in computer vision. However, the pre-training of MIM alw…

AttributeRepresentation LearningSelf-Supervised Learning

FreMIM: Fourier Transform Meets Masked Image Modeling for Medical Image Segmentation

2023-04-21 · Wenxuan Wang, Jing Wang, Chen Chen, Jianbo Jiao 외

The research community has witnessed the powerful potential of self-supervised Masked Image Modeling (MIM), which enables the models capable of learning visual representation from unlabeled data. In this paper, to incorp…

Image SegmentationMedical Image SegmentationRepresentation LearningSemantic Segmentation

Self-Supervised Learning for Visual Relationship Detection through Masked Bounding Box Reconstruction

2023-11-08 · Zacharias Anastasakis, Dimitrios Mallis, Markos Diomataris, George Alexandridis 외

We present a novel self-supervised approach for representation learning, particularly for the task of Visual Relationship Detection (VRD). Motivated by the effectiveness of Masked Image Modeling (MIM), we propose Masked …

Predicate DetectionRelationship DetectionRepresentation LearningSelf-Supervised Learning+1

Unsupervised Anomaly Detection in Medical Images Using Masked Diffusion Model

2023-05-31 · Hasan Iqbal, Umar Khalid, Jing Hua, Chen Chen

It can be challenging to identify brain MRI anomalies using supervised deep-learning techniques due to anatomical heterogeneity and the requirement for pixel-level labeling. Unsupervised anomaly detection approaches prov…

AnatomyAnomaly DetectionUnsupervised Anomaly Detection