在本文中，我们提出了Mixed and Masked AutoEncoder（MixMAE），这是一种简单但高效的预训练方法，适用于各种分层Vision Transformers。现有的用于分层Vision Transformers的masked image modeling（MIM）方法用特殊的[MASK]符号替换输入token的随机子集，旨在从损坏的图像中重建原始图像token。然而，我们发现使用[MASK]符号会大大减慢训练速度，并由于较大的掩码比例（例如SimMIM中的60%）导致预训练与微调不一致。另一方面，MAE在其编码器中完全不引入[MASK]token，但不适用于分层Vision Transformers。为了解决该问题并加速分层模型的预训练，我们将一张图像的掩码token替换为另一张图像的可见token，即创建一张混合图像。然后，我们进行双重重建，从混合输入中重建两张原始图像，这显著提高了效率。虽然MixMAE可应用于各种分层Transformers，但本文探索使用具有大窗口尺寸的Swin Transformer，并将其扩展到巨大的模型尺寸（达到600M参数）。实验结果表明，MixMAE能够高效地学习高质量的视觉表示。值得注意的是，采用Swin-B/W14的MixMAE在ImageNet-1K上经过600个epoch的预训练后，达到了85.1%的top-1准确率。此外，它在其他6个数据集上的迁移性能表明，与以往流行的MIM方法相比，MixMAE具有更好的FLOPs/性能权衡。代码可在https://github.com/Sense-X/MixMIM获取。
英文摘要：
In this paper, we propose Mixed and Masked AutoEncoder (MixMAE), a simple but efficient pretraining method that is applicable to various hierarchical Vision Transformers. Existing masked image modeling (MIM) methods for hierarchical Vision Transformers replace a random subset of input tokens with a special [MASK] symbol and aim at reconstructing original image tokens from the corrupted image. However, we find that using the [MASK] symbol greatly slows down the training and causes pretraining-finetuning inconsistency, due to the large masking ratio (e.g., 60% in SimMIM). On the other hand, MAE does not introduce [MASK] tokens at its encoder at all but is not applicable for hierarchical Vision Transformers. To solve the issue and accelerate the pretraining of hierarchical models, we replace the masked tokens of one image with visible tokens of another image, i.e., creating a mixed image. We then conduct dual reconstruction to reconstruct the two original images from the mixed input, which significantly improves efficiency. While MixMAE can be applied to various hierarchical Transformers, this paper explores using Swin Transformer with a large window size and scales up to huge model size (to reach 600M parameters). Empirical results demonstrate that MixMAE can learn high-quality visual representations efficiently. Notably, MixMAE with Swin-B/W14 achieves 85.1% top-1 accuracy on ImageNet-1K by pretraining for 600 epochs. Besides, its transfer performances on the other 6 datasets show that MixMAE has better FLOPs / performance tradeoff than previous popular MIM methods. Code is available at https://github.com/Sense-X/MixMIM.
The Chinese University of Hong Kong（香港中文大学）；NVIDIA AI Technology Center（英伟达人工智能技术中心）；SenseTime Research（商汤科技研究院）；Tsinghua University（清华大学）
FlowFormer introduces a transformer architecture into optical flow estimation and achieves state-of-the-art performance. The core component of FlowFormer is the transformer-based cost-volume encoder. Inspired by the recent success of masked autoencoding (MAE) pretraining in unleashing transformers' capacity of encoding visual representation, we propose Masked Cost Volume Autoencoding (MCVA) to enhance FlowFormer by pretraining the cost-volume encoder with a novel MAE scheme. Firstly, we introduce a block-sharing masking strategy to prevent masked information leakage, as the cost maps of neighb…
Diffusion Masked Pretraining for Dynamic Point Cloud
Xi’an Jiaotong University（西安交通大学）；School of Artificial Intelligence and Robotics, Hunan University（湖南大学人工智能与机器人学院）；University of Western Australia（西澳大学）
Efficient Self-supervised Vision Pretraining with Local Masked Reconstruction
King Abdullah University of Science and Technology（阿卜杜拉国王科技大学）；Monash University（莫纳什大学）；Nile University（尼罗河大学）；Nanyang Technological University（南洋理工大学）
Semantics-Guided Multimodal Masked Autoencoder Pretraining for 3D BEV Object Detection
University of Lincoln, Lincoln Centre for Autonomous Systems（林肯大学，林肯自主系统中心）；University of Cambridge, Institute for Manufacturing, Department of Engineering（剑桥大学，制造研究所，工程系）
Understanding Self-Supervised Pretraining with Part-Aware Representation Learning
Key Lab of High Confidence Software Technologies (Peking University), Ministry of Education, China（高可信软件技术教育部重点实验室（北京大学））；School of Computer Science, Peking University（北京大学计算机学院）；School of AI, Peking University（北京大学人工智能学院）；University of Hong Kong（香港大学）；School of EIC, Huazhong University of Science & Technology（华中科技大学电子信息与通信学院）；Baidu（百度）；UC Berkeley（加州大学伯克利分校）
Masked autoencoding has achieved great success for self-supervised learning in the image and language domains. However, mask based pretraining has yet to show benefits for point cloud understanding, likely due to standard backbones like PointNet being unable to properly handle the training versus testing distribution mismatch introduced by masking during training. In this paper, we bridge this gap by proposing a discriminative mask pretraining Transformer framework, MaskPoint}, for point clouds. Our key idea is to represent the point cloud as discrete occupancy values (1 if part of the point c…
Deblurring Masked Autoencoder is Better Recipe for Ultrasound Image Recognition
West China Hospital, Sichuan University（四川大学华西医院）；Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）；Sichuan University（四川大学）；Beijing University of Posts and Telecommunications（北京邮电大学）
We introduce a vision-language foundation model called VL-BEiT, which is a bidirectional multimodal Transformer learned by generative pretraining. Our minimalist solution conducts masked prediction on both monomodal and multimodal data with a shared Transformer. Specifically, we perform masked vision-language modeling on image-text pairs, masked language modeling on texts, and masked image modeling on images. VL-BEiT is learned from scratch with one unified pretraining task, one shared backbone, and one-stage training. Our method is conceptually simple and empirically effective. Experimental r…
Shanghai AI Laboratory（上海人工智能实验室）；SenseTime Research（商汤科技研究院）
Vision Transformers (ViT) become widely-adopted architectures for various vision tasks. Masked auto-encoding for feature pretraining and multi-scale hybrid convolution-transformer architectures can further unleash the potentials of ViT, leading to state-of-the-art performances on image classification, detection and semantic segmentation. In this paper, our ConvMAE framework demonstrates that multi-scale hybrid convolution-transformer can learn more discriminative representations via the mask auto-encoding scheme. However, directly using the original masking strategy leads to the heavy computat…
Motivated by the fact that forward and backward passes of a deep network naturally form symmetric mappings between input and output representations, we introduce a simple yet effective self-supervised vision model pretraining framework inspired by energy-based models (EBMs). In the proposed framework, we model energy estimation and data restoration as the forward and backward passes of a single network without any auxiliary components, e.g., an extra decoder. For the forward pass, we fit a network to an energy function that assigns low energy scores to samples that belong to an unlabeled datas…
University of Science and Technology of China（中国科学技术大学）；Microsoft Research Asia（微软亚洲研究院）
We propose bootstrapped masked autoencoders (BootMAE), a new approach for vision BERT pretraining. BootMAE improves the original masked autoencoders (MAE) with two core designs: 1) momentum encoder that provides online feature as extra BERT prediction targets; 2) target-aware decoder that tries to reduce the pressure on the encoder to memorize target-specific information in BERT pretraining. The first design is motivated by the observation that using a pretrained MAE to extract the features as the BERT prediction target for masked tokens can achieve better pretraining performance. Therefore, w…
Emory University（埃默里大学）；University of Southern California（南加州大学）；Arizona State University（亚利桑那州立大学）；Mayo Clinic（妙佑医疗国际）
Language modality within the vision language pretraining framework is innately discretized, endowing each word in the language vocabulary a semantic meaning. In contrast, visual modality is inherently continuous and high-dimensional, which potentially prohibits the alignment as well as fusion between vision and language modalities. We therefore propose to "discretize" the visual representation by joint learning a codebook that imbues each visual token a semantic. We then utilize these discretized visual semantics as self-supervised ground-truths for building our Masked Image Modeling objective…
Self-attention based transformer models have been dominating many computer vision tasks in the past few years. Their superb model qualities heavily depend on the excessively large labeled image datasets. In order to reduce the reliance on large labeled datasets, reconstruction based masked autoencoders are gaining popularity, which learn high quality transferable representations from unlabeled images. For the same purpose, recent weakly supervised image pretraining methods explore language supervision from text captions accompanying the images. In this work, we propose masked image pretraining…
$\mathsf{CSMAE~}$:~Cataract Surgical Masked Autoencoder (MAE) based Pre-training
Johns Hopkins University（约翰斯·霍普金斯大学）；Wilmer Eye Institute, Johns Hopkins University（约翰斯·霍普金斯大学威尔默眼科研究所）；Malone Center for Engineering in Healthcare, Johns Hopkins University（约翰斯·霍普金斯大学马龙医疗保健工程中心）
Better with Less: Tackling Heterogeneous Multi-Modal Image Joint Pretraining via Conditioned and Degraded Masked Autoencoder
College of Electronic Science and Technology, National University of Defense Technology (NUDT)（电子科学与技术学院，国防科技大学）；Beijing Institute of Remote Sensing Information（遥感信息研究所）
University of Surrey（萨里大学）；ByteDance（字节跳动）；The University of Hong Kong（香港大学）；Harbin Institute of Technology（哈尔滨工业大学）
Masked autoencoders (MAEs) have emerged recently as art self-supervised spatiotemporal representation learners. Inheriting from the image counterparts, however, existing video MAEs still focus largely on static appearance learning whilst are limited in learning dynamic temporal information hence less effective for video downstream tasks. To resolve this drawback, in this work we present a motion-aware variant -- MotionMAE. Apart from learning to reconstruct individual masked patches of video frames, our model is designed to additionally predict the corresponding motion structure information ov…
Tissue-Contrastive Semi-Masked Autoencoders for Segmentation Pretraining on Chest CT
Hangzhou City University（杭州城市大学）；Southwest Hospital, Army Medical University (Third Military Medical University)（陆军军医大学（第三军医大学）西南医院）；Zhejiang University（浙江大学）
Beyond ViT Tokens: Masked-Diffusion Pretrained Convolutional Pathology Foundation Model for Cell-Level Dense Prediction
Tsinghua Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生院，清华大学）；Research Institute of Tsinghua, Pearl River Delta（清华大学 Pearl River Delta 研究院）；The Chinese University of Hong Kong, ShenZhen（香港中文大学（深圳））
University of Chinese Academy of Sciences（中国科学院大学）；Microsoft Research（微软研究院）
Masked image modeling has demonstrated great potential to eliminate the label-hungry problem of training large-scale vision Transformers, achieving impressive performance on various downstream tasks. In this work, we propose a unified view of masked image modeling after revisiting existing methods. Under the unified view, we introduce a simple yet effective method, termed as MaskDistill, which reconstructs normalized semantic features from teacher models at the masked positions, conditioning on corrupted input images. Experimental results on image classification and semantic segmentation show …
We introduce a pretraining technique called Selfie, which stands for SELFie supervised Image Embedding. Selfie generalizes the concept of masked language modeling of BERT (Devlin et al., 2019) to continuous data, such as images, by making use of the Contrastive Predictive Coding loss (Oord et al., 2018). Given masked-out patches in an input image, our method learns to select the correct patch, among other "distractor" patches sampled from the same image, to fill in the masked location. This classification objective sidesteps the need for predicting exact pixel values of the target patches. The…
Medical image visual question answering (VQA) is a task to answer clinical questions, given a radiographic image, which is a challenging problem that requires a model to integrate both vision and language information. To solve medical VQA problems with a limited number of training data, pretrain-finetune paradigm is widely used to improve the model generalization. In this paper, we propose a self-supervised method that applies Masked image modeling, Masked language modeling, Image text matching and Image text alignment via contrastive learning (M2I2) for pretraining on medical image caption da…
Scene text recognition (STR) enables computers to recognize and read the text in various real-world scenes. Recent STR models benefit from taking linguistic information in addition to visual cues into consideration. We propose a novel Masked Vision-Language Transformers (MVLT) to capture both the explicit and the implicit linguistic information. Our encoder is a Vision Transformer, and our decoder is a multi-modal Transformer. MVLT is trained in two stages: in the first stage, we design a STR-tailored pretraining method based on a masking strategy; in the second stage, we fine-tune our model a…
