Search

Vision-and-Language Navigation (VLN) aims to develop embodied agents that navigate based on human instructions. However, current VLN frameworks often rely on static environments and optimal expert supervision, limiting their real-world applicability. To address this, we introduce Human-Aware Vision-and-Language Navigation (HA-VLN), extending traditional VLN by incorporating dynamic human activities and relaxing key assumptions. We propose the Human-Aware 3D (HA3D) simulator, which combines dynamic human activities with the Matterport3D dataset, and the Human-Aware Room-to-Room (HA-R2R) dataset, extending R2R with human activity descriptions. To tackle HA-VLN challenges, we present the Expert-Supervised Cross-Modal (VLN-CM) and Non-Expert-Supervised Decision Transformer (VLN-DT) agents, utilizing cross-modal fusion and diverse training strategies for effective navigation in dynamic human environments. A comprehensive evaluation, including metrics considering human activities, and systematic analysis of HA-VLN's unique challenges, underscores the need for further research to enhance HA-VLN agents' real-world robustness and adaptability. Ultimately, this work provides benchmarks and insights for future research on embodied AI and Sim2Real transfer, paving the way for more realistic and applicable VLN systems in human-populated environments., Comment: 30 pages, 18 figures, Project Page: https://lpercc.github.io/HA3D_simulator/

Modern vision models are trained on very large noisy datasets. While these models acquire strong capabilities, they may not follow the user's intent to output the desired results in certain aspects, e.g., visual aesthetic, preferred style, and responsibility. In this paper, we target the realm of visual aesthetics and aim to align vision models with human aesthetic standards in a retrieval system. Advanced retrieval systems usually adopt a cascade of aesthetic models as re-rankers or filters, which are limited to low-level features like saturation and perform poorly when stylistic, cultural or knowledge contexts are involved. We find that utilizing the reasoning ability of large language models (LLMs) to rephrase the search query and extend the aesthetic expectations can make up for this shortcoming. Based on the above findings, we propose a preference-based reinforcement learning method that fine-tunes the vision models to distill the knowledge from both LLMs reasoning and the aesthetic models to better align the vision models with human aesthetics. Meanwhile, with rare benchmarks designed for evaluating retrieval systems, we leverage large multi-modality model (LMM) to evaluate the aesthetic performance with their strong abilities. As aesthetic assessment is one of the most subjective tasks, to validate the robustness of LMM, we further propose a novel dataset named HPIR to benchmark the alignment with human aesthetics. Experiments demonstrate that our method significantly enhances the aesthetic behaviors of the vision models, under several metrics. We believe the proposed algorithm can be a general practice for aligning vision models with human values., Comment: 28 pages, 26 figures, under review

Text-guided video prediction (TVP) involves predicting the motion of future frames from the initial frame according to an instruction, which has wide applications in virtual reality, robotics, and content creation. Previous TVP methods make significant breakthroughs by adapting Stable Diffusion for this task. However, they struggle with frame consistency and temporal stability primarily due to the limited scale of video datasets. We observe that pretrained Image2Video diffusion models possess good priors for video dynamics but they lack textual control. Hence, transferring Image2Video models to leverage their video dynamic priors while injecting instruction control to generate controllable videos is both a meaningful and challenging task. To achieve this, we introduce the Multi-Modal Large Language Model (MLLM) to predict future video states based on initial frames and text instructions. More specifically, we design a dual query transformer (DQFormer) architecture, which integrates the instructions and frames into the conditional embeddings for future frame prediction. Additionally, we develop Long-Short Term Temporal Adapters and Spatial Adapters that can quickly transfer general video diffusion models to specific scenarios with minimal training costs. Experimental results show that our method significantly outperforms state-of-the-art techniques on four datasets: Something Something V2, Epic Kitchen-100, Bridge Data, and UCF-101. Notably, AID achieves 91.2% and 55.5% FVD improvements on Bridge and SSv2 respectively, demonstrating its effectiveness in various domains. More examples can be found at our website https://chenhsing.github.io/AID.

Despite impressive advancements in diffusion-based video editing models in altering video attributes, there has been limited exploration into modifying motion information while preserving the original protagonist's appearance and background. In this paper, we propose MotionFollower, a lightweight score-guided diffusion model for video motion editing. To introduce conditional controls to the denoising process, MotionFollower leverages two of our proposed lightweight signal controllers, one for poses and the other for appearances, both of which consist of convolution blocks without involving heavy attention calculations. Further, we design a score guidance principle based on a two-branch architecture, including the reconstruction and editing branches, which significantly enhance the modeling capability of texture details and complicated backgrounds. Concretely, we enforce several consistency regularizers and losses during the score estimation. The resulting gradients thus inject appropriate guidance to the intermediate latents, forcing the model to preserve the original background details and protagonists' appearances without interfering with the motion modification. Experiments demonstrate the competitive motion editing ability of MotionFollower qualitatively and quantitatively. Compared with MotionEditor, the most advanced motion editing model, MotionFollower achieves an approximately 80% reduction in GPU memory while delivering superior motion editing performance and exclusively supporting large camera movements and actions., Comment: 23 pages, 18 figures. Project page at https://francis-rings.github.io/MotionFollower/

Objectives: To assess the effectiveness of digital scanning techniques for self-assessment and of preparations and restorations in preclinical dental education when compared to traditional faculty grading. Methods: Forty-four separate Class I (#30-O), Class II (#30-MO) preparations, and class II amalgam restorations (#31-MO) were generated respectively under preclinical assessment setting. Calibrated faculty evaluated the preparations and restorations using a standard rubric from preclinical operative class. The same teeth were scanned using Planmeca PlanScan intraoral scanner and graded using the Romexis E4D Compare Software. Each tooth was compared against a corresponding gold standard tooth with tolerance intervals ranging from 100{\mu}m to 500{\mu}m. These scores were compared to traditional faculty grades using a linear mixed model to estimate the mean differences at 95% confidence interval for each tolerance level. Results: The average Compare Software grade of Class I preparation at 300{\mu}m tolerance had the smallest mean difference of 1.64 points on a 100 points scale compared to the average faculty grade. Class II preparation at 400{\mu}m tolerance had the smallest mean difference of 0.41 points. Finally, Class II Restoration at 300{\mu}m tolerance had the smallest mean difference at 0.20 points. Conclusion: In this study, tolerance levels that best correlated the Compare Software grades with the faculty grades were determined for three operative procedures: class I preparation, class II preparation and class II restoration. This Compare Software can be used as a useful adjunct method for more objective grading. It also can be used by students as a great self-assessment tool.

Automatic measurements of tear meniscus height (TMH) have been achieved by using deep learning techniques; however, annotation is significantly influenced by subjective factors and is both time-consuming and labor-intensive. In this paper, we introduce an automatic TMH measurement technique based on edge detection-assisted annotation within a deep learning framework. This method generates mask labels less affected by subjective factors with enhanced efficiency compared to previous annotation approaches. For improved segmentation of the pupil and tear meniscus areas, the convolutional neural network Inceptionv3 was first implemented as an image quality assessment model, effectively identifying higher-quality images with an accuracy of 98.224%. Subsequently, by using the generated labels, various algorithms, including Unet, ResUnet, Deeplabv3+FcnResnet101, Deeplabv3+FcnResnet50, FcnResnet50, and FcnResnet101 were trained, with Unet demonstrating the best performance. Finally, Unet was used for automatic pupil and tear meniscus segmentation to locate the center of the pupil and calculate TMH,respectively. An evaluation of the mask quality predicted by Unet indicated a Mean Intersection over Union of 0.9362, a recall of 0.9261, a precision of 0.9423, and an F1-Score of 0.9326. Additionally, the TMH predicted by the model was assessed, with the fitting curve represented as y= 0.982x-0.862, an overall correlation coefficient of r^2=0.961 , and an accuracy of 94.80% (237/250). In summary, the algorithm can automatically screen images based on their quality,segment the pupil and tear meniscus areas, and automatically measure TMH. Measurement results using the AI algorithm demonstrate a high level of consistency with manual measurements, offering significant support to clinical doctors in diagnosing dry eye disease., Comment: 22 pages, 5 figures

In this paper, we explore the localization features of wave functions in a family of mosaic quasiperiodic chains obtained by continuously interpolating between two limits: the mosaic Aubry-Andr\'{e} (AA) model, known for its exact mobility edges with extended states in the band-center region, and localized ones in the band-edge regions for a large enough modulation amplitude, and the mosaic Fibonacci chain, which exhibits its multifractal nature for all the states except for the extended one with $E=0$ for an arbitrary finite modulation amplitude. We discover that the mosaic AA limit for the states in the band-edge regions evolves into multifractal ones through a cascade of delocalization transitions. This cascade shows lobes of lower fractal dimension values separated by maxima of fractal dimension. In contrast, the states in the band-center region (except for the $E=0$ state) display an anomalous cascading process, where it emerges lobes of higher fractal dimension values are separated by the regions with lower fractal dimensions. Our findings offer insight into understanding the multifractality of quasiperiodic chains., Comment: 12 pages, 11 figures

The recent wave of AI-generated content (AIGC) has witnessed substantial success in computer vision, with the diffusion model playing a crucial role in this achievement. Due to their impressive generative capabilities, diffusion models are gradually superseding methods based on GANs and auto-regressive Transformers, demonstrating exceptional performance not only in image generation and editing, but also in the realm of video-related research. However, existing surveys mainly focus on diffusion models in the context of image generation, with few up-to-date reviews on their application in the video domain. To address this gap, this paper presents a comprehensive review of video diffusion models in the AIGC era. Specifically, we begin with a concise introduction to the fundamentals and evolution of diffusion models. Subsequently, we present an overview of research on diffusion models in the video domain, categorizing the work into three key areas: video generation, video editing, and other video understanding tasks. We conduct a thorough review of the literature in these three key areas, including further categorization and practical contributions in the field. Finally, we discuss the challenges faced by research in this domain and outline potential future developmental trends. A comprehensive list of video diffusion models studied in this survey is available at https://github.com/ChenHsing/Awesome-Video-Diffusion-Models.

We extend the discussion on the difference between angular momentum and pseudo-angular momentum in field theory. We show that the often quoted expressions in [Phys.Rev.B 103, L100409 (2021)] only apply to a non-linear system, and derive the correct rotation symmetry and the corresponding angular momentum for a linear elastic system governed by Navier-Cauchy equation. By mapping the concepts and methods for the elastic wave into electromagnetic theory, we argue that the renowned canonical and Benlinfante angular momentum of light are actually pseudo-angular momentum. Then, we derive the ``Newtonian" momentum $\int \text{d}^3 x\boldsymbol{E}$ and angular momentum $\int \text{d}^3 x (\boldsymbol{r}\times\boldsymbol{E})$ for a free electromagnetic wave, which are conserved quantities during propagation in vacuum., Comment: 8 pages, no figures

The recent wave of AI-generated content has witnessed the great development and success of Text-to-Image (T2I) technologies. By contrast, Text-to-Video (T2V) still falls short of expectations though attracting increasing interests. Existing works either train from scratch or adapt large T2I model to videos, both of which are computation and resource expensive. In this work, we propose a Simple Diffusion Adapter (SimDA) that fine-tunes only 24M out of 1.1B parameters of a strong T2I model, adapting it to video generation in a parameter-efficient way. In particular, we turn the T2I model for T2V by designing light-weight spatial and temporal adapters for transfer learning. Besides, we change the original spatial attention to the proposed Latent-Shift Attention (LSA) for temporal consistency. With similar model architecture, we further train a video super-resolution model to generate high-definition (1024x1024) videos. In addition to T2V generation in the wild, SimDA could also be utilized in one-shot video editing with only 2 minutes tuning. Doing so, our method could minimize the training effort with extremely few tunable parameters for model adaptation.

Contrastive language-image pretraining (CLIP) has demonstrated remarkable success in various image tasks. However, how to extend CLIP with effective temporal modeling is still an open and crucial problem. Existing factorized or joint spatial-temporal modeling trades off between the efficiency and performance. While modeling temporal information within straight through tube is widely adopted in literature, we find that simple frame alignment already provides enough essence without temporal attention. To this end, in this paper, we proposed a novel Implicit Learnable Alignment (ILA) method, which minimizes the temporal modeling effort while achieving incredibly high performance. Specifically, for a frame pair, an interactive point is predicted in each frame, serving as a mutual information rich region. By enhancing the features around the interactive point, two frames are implicitly aligned. The aligned features are then pooled into a single token, which is leveraged in the subsequent spatial self-attention. Our method allows eliminating the costly or insufficient temporal self-attention in video. Extensive experiments on benchmarks demonstrate the superiority and generality of our module. Particularly, the proposed ILA achieves a top-1 accuracy of 88.7% on Kinetics-400 with much fewer FLOPs compared with Swin-L and ViViT-H. Code is released at https://github.com/Francis-Rings/ILA ., Comment: ICCV 2023 oral. 14 pages, 7 figures. Code released at https://github.com/Francis-Rings/ILA

Charts are a powerful tool for visually conveying complex data, but their comprehension poses a challenge due to the diverse chart types and intricate components. Existing chart comprehension methods suffer from either heuristic rules or an over-reliance on OCR systems, resulting in suboptimal performance. To address these issues, we present ChartReader, a unified framework that seamlessly integrates chart derendering and comprehension tasks. Our approach includes a transformer-based chart component detection module and an extended pre-trained vision-language model for chart-to-X tasks. By learning the rules of charts automatically from annotated datasets, our approach eliminates the need for manual rule-making, reducing effort and enhancing accuracy.~We also introduce a data variable replacement technique and extend the input and position embeddings of the pre-trained model for cross-task training. We evaluate ChartReader on Chart-to-Table, ChartQA, and Chart-to-Text tasks, demonstrating its superiority over existing methods. Our proposed framework can significantly reduce the manual effort involved in chart analysis, providing a step towards a universal chart understanding model. Moreover, our approach offers opportunities for plug-and-play integration with mainstream LLMs such as T5 and TaPas, extending their capability to chart comprehension tasks. The code is available at https://github.com/zhiqic/ChartReader.

In order to reveal the rationale behind model predictions, many works have exploited providing explanations in various forms. Recently, to further guarantee readability, more and more works turn to generate sentence-level human language explanations. However, current works pursuing sentence-level explanations rely heavily on annotated training data, which limits the development of interpretability to only a few tasks. As far as we know, this paper is the first to explore this problem smoothly from weak-supervised learning to unsupervised learning. Besides, we also notice the high latency of autoregressive sentence-level explanation generation, which leads to asynchronous interpretability after prediction. Therefore, we propose a non-autoregressive interpretable model to facilitate parallel explanation generation and simultaneous prediction. Through extensive experiments on Natural Language Inference task and Spouse Prediction task, we find that users are able to train classifiers with comparable performance $10-15\times$ faster with parallel explanation generation using only a few or no annotated training data., Comment: Accepted by ICASSP 2023

Unlike language tasks, where the output space is usually limited to a set of tokens, the output space of visual tasks is more complicated, making it difficult to build a unified visual model for various visual tasks. In this paper, we seek to unify the output space of visual tasks, so that we can also build a unified model for visual tasks. To this end, we demonstrate a single unified model that simultaneously handles two typical visual tasks of instance segmentation and depth estimation, which have discrete/fixed-length and continuous/varied-length outputs, respectively. We propose several new techniques that take into account the particularity of visual tasks: 1) Soft token. We employ soft token to represent the task output. Unlike hard tokens in the common VQ-VAE which are assigned one-hot to discrete codebooks/vocabularies, the soft token is assigned softly to the codebook embeddings. Soft token can improve the accuracy of both the next token inference and decoding of the task output; 2) Mask augmentation. Many visual tasks have corruption, undefined or invalid values in label annotations, i.e., occluded area of depth maps. We show that a mask augmentation technique can greatly benefit these tasks. With these new techniques and other designs, we show that the proposed general-purpose task-solver can perform both instance segmentation and depth estimation well. Particularly, we achieve 0.279 RMSE on the specific task of NYUv2 depth estimation, setting a new record on this benchmark. The general-purpose task-solver, dubbed AiT, is available at \url{https://github.com/SwinTransformer/AiT}.

Vision Transformers (ViTs) have achieved overwhelming success, yet they suffer from vulnerable resolution scalability, i.e., the performance drops drastically when presented with input resolutions that are unseen during training. We introduce, ResFormer, a framework that is built upon the seminal idea of multi-resolution training for improved performance on a wide spectrum of, mostly unseen, testing resolutions. In particular, ResFormer operates on replicated images of different resolutions and enforces a scale consistency loss to engage interactive information across different scales. More importantly, to alternate among varying resolutions effectively, especially novel ones in testing, we propose a global-local positional embedding strategy that changes smoothly conditioned on input sizes. We conduct extensive experiments for image classification on ImageNet. The results provide strong quantitative evidence that ResFormer has promising scaling abilities towards a wide range of resolutions. For instance, ResFormer-B-MR achieves a Top-1 accuracy of 75.86% and 81.72% when evaluated on relatively low and high resolutions respectively (i.e., 96 and 640), which are 48% and 7.49% better than DeiT-B. We also demonstrate, moreover, ResFormer is flexible and can be easily extended to semantic segmentation, object detection and video action recognition. Code is available at https://github.com/ruitian12/resformer., Comment: CVPR 2023

Semi-supervised action recognition is a challenging but critical task due to the high cost of video annotations. Existing approaches mainly use convolutional neural networks, yet current revolutionary vision transformer models have been less explored. In this paper, we investigate the use of transformer models under the SSL setting for action recognition. To this end, we introduce SVFormer, which adopts a steady pseudo-labeling framework (ie, EMA-Teacher) to cope with unlabeled video samples. While a wide range of data augmentations have been shown effective for semi-supervised image classification, they generally produce limited results for video recognition. We therefore introduce a novel augmentation strategy, Tube TokenMix, tailored for video data where video clips are mixed via a mask with consistent masked tokens over the temporal axis. In addition, we propose a temporal warping augmentation to cover the complex temporal variation in videos, which stretches selected frames to various temporal durations in the clip. Extensive experiments on three datasets Kinetics-400, UCF-101, and HMDB-51 verify the advantage of SVFormer. In particular, SVFormer outperforms the state-of-the-art by 31.5% with fewer training epochs under the 1% labeling rate of Kinetics-400. Our method can hopefully serve as a strong benchmark and encourage future search on semi-supervised action recognition with Transformer networks.

Grounded Situation Recognition (GSR) aims to generate structured semantic summaries of images for "human-like" event understanding. Specifically, GSR task not only detects the salient activity verb (e.g. buying), but also predicts all corresponding semantic roles (e.g. agent and goods). Inspired by object detection and image captioning tasks, existing methods typically employ a two-stage framework: 1) detect the activity verb, and then 2) predict semantic roles based on the detected verb. Obviously, this illogical framework constitutes a huge obstacle to semantic understanding. First, pre-detecting verbs solely without semantic roles inevitably fails to distinguish many similar daily activities (e.g., offering and giving, buying and selling). Second, predicting semantic roles in a closed auto-regressive manner can hardly exploit the semantic relations among the verb and roles. To this end, in this paper we propose a novel two-stage framework that focuses on utilizing such bidirectional relations within verbs and roles. In the first stage, instead of pre-detecting the verb, we postpone the detection step and assume a pseudo label, where an intermediate representation for each corresponding semantic role is learned from images. In the second stage, we exploit transformer layers to unearth the potential semantic relations within both verbs and semantic roles. With the help of a set of support images, an alternate learning scheme is designed to simultaneously optimize the results: update the verb using nouns corresponding to the image, and update nouns using verbs from support images. Extensive experimental results on challenging SWiG benchmarks show that our renovated framework outperforms other state-of-the-art methods under various metrics., Comment: ACM Multimedia 2022 (Oral), Code: https://github.com/zhiqic/GSRFormer

Previous work generally believes that improving the spatial invariance of convolutional networks is the key to object counting. However, after verifying several mainstream counting networks, we surprisingly found too strict pixel-level spatial invariance would cause overfit noise in the density map generation. In this paper, we try to use locally connected Gaussian kernels to replace the original convolution filter to estimate the spatial position in the density map. The purpose of this is to allow the feature extraction process to potentially stimulate the density map generation process to overcome the annotation noise. Inspired by previous work, we propose a low-rank approximation accompanied with translation invariance to favorably implement the approximation of massive Gaussian convolution. Our work points a new direction for follow-up research, which should investigate how to properly relax the overly strict pixel-level spatial invariance for object counting. We evaluate our methods on 4 mainstream object counting networks (i.e., MCNN, CSRNet, SANet, and ResNet-50). Extensive experiments were conducted on 7 popular benchmarks for 3 applications (i.e., crowd, vehicle, and plant counting). Experimental results show that our methods significantly outperform other state-of-the-art methods and achieve promising learning of the spatial position of objects., Comment: Accepted to CVPR 2022, Code: https://github.com/zhiqic/Rethinking-Counting

An important goal of self-supervised learning is to enable model pre-training to benefit from almost unlimited data. However, one method that has recently become popular, namely masked image modeling (MIM), is suspected to be unable to benefit from larger data. In this work, we break this misconception through extensive experiments, with data scales ranging from 10\% of ImageNet-1K to full ImageNet-22K, model sizes ranging from 49 million to 1 billion, and training lengths ranging from 125K iterations to 500K iterations. Our study reveals that: (i) Masked image modeling is also demanding on larger data. We observed that very large models got over-fitted with relatively small data; (ii) The length of training matters. Large models trained with masked image modeling can benefit from more data with longer training; (iii) The validation loss in pre-training is a good indicator to measure how well the model performs for fine-tuning on multiple tasks. This observation allows us to pre-evaluate pre-trained models in advance without having to make costly trial-and-error assessments of downstream tasks. We hope that our findings will advance the understanding of masked image modeling in terms of scaling ability.

Recently, masked image modeling (MIM) has offered a new methodology of self-supervised pre-training of vision transformers. A key idea of efficient implementation is to discard the masked image patches (or tokens) throughout the target network (encoder), which requires the encoder to be a plain vision transformer (e.g., ViT), albeit hierarchical vision transformers (e.g., Swin Transformer) have potentially better properties in formulating vision inputs. In this paper, we offer a new design of hierarchical vision transformers named HiViT (short for Hierarchical ViT) that enjoys both high efficiency and good performance in MIM. The key is to remove the unnecessary "local inter-unit operations", deriving structurally simple hierarchical vision transformers in which mask-units can be serialized like plain vision transformers. For this purpose, we start with Swin Transformer and (i) set the masking unit size to be the token size in the main stage of Swin Transformer, (ii) switch off inter-unit self-attentions before the main stage, and (iii) eliminate all operations after the main stage. Empirical studies demonstrate the advantageous performance of HiViT in terms of fully-supervised, self-supervised, and transfer learning. In particular, in running MAE on ImageNet-1K, HiViT-B reports a +0.6% accuracy gain over ViT-B and a 1.9$\times$ speed-up over Swin-B, and the performance gain generalizes to downstream tasks of detection and segmentation. Code will be made publicly available.

With Vision Transformers (ViTs) making great advances in a variety of computer vision tasks, recent literature have proposed various variants of vanilla ViTs to achieve better efficiency and efficacy. However, it remains unclear how their unique architecture impact robustness towards common corruptions. In this paper, we make the first attempt to probe into the robustness gap among ViT variants and explore underlying designs that are essential for robustness. Through an extensive and rigorous benchmarking, we demonstrate that simple architecture designs such as overlapping patch embedding and convolutional feed-forward network (FFN) can promote the robustness of ViTs. Moreover, since training ViTs relies heavily on data augmentation, whether previous CNN-based augmentation strategies that are targeted at robustness purposes can still be useful is worth investigating. We explore different data augmentation on ViTs and verify that adversarial noise training is powerful while fourier-domain augmentation is inferior. Based on these findings, we introduce a novel conditional method of generating dynamic augmentation parameters conditioned on input images, offering state-of-the-art robustness towards common corruptions.

The imbalanced classification problem turns out to be one of the important and challenging problems in data mining and machine learning. The performances of traditional classifiers will be severely affected by many data problems, such as class imbalanced problem, class overlap and noise. The Tomek-Link algorithm was only used to clean data when it was proposed. In recent years, there have been reports of combining Tomek-Link algorithm with sampling technique. The Tomek-Link sampling algorithm can effectively reduce the class overlap on data, remove the majority instances that are difficult to distinguish, and improve the algorithm classification accuracy. However, the Tomek-Links under-sampling algorithm only considers the boundary instances that are the nearest neighbors to each other globally and ignores the potential local overlapping instances. When the number of minority instances is small, the under-sampling effect is not satisfactory, and the performance improvement of the classification model is not obvious. Therefore, on the basis of Tomek-Link, a multi-granularity relabeled under-sampling algorithm (MGRU) is proposed. This algorithm fully considers the local information of the data set in the local granularity subspace, and detects the local potential overlapping instances in the data set. Then, the overlapped majority instances are eliminated according to the global relabeled index value, which effectively expands the detection range of Tomek-Links. The simulation results show that when we select the optimal global relabeled index value for under-sampling, the classification accuracy and generalization performance of the proposed under-sampling algorithm are significantly better than other baseline algorithms., Comment: 35 pages

This paper presents SimMIM, a simple framework for masked image modeling. We simplify recently proposed related approaches without special designs such as block-wise masking and tokenization via discrete VAE or clustering. To study what let the masked image modeling task learn good representations, we systematically study the major components in our framework, and find that simple designs of each component have revealed very strong representation learning performance: 1) random masking of the input image with a moderately large masked patch size (e.g., 32) makes a strong pre-text task; 2) predicting raw pixels of RGB values by direct regression performs no worse than the patch classification approaches with complex designs; 3) the prediction head can be as light as a linear layer, with no worse performance than heavier ones. Using ViT-B, our approach achieves 83.8% top-1 fine-tuning accuracy on ImageNet-1K by pre-training also on this dataset, surpassing previous best approach by +0.6%. When applied on a larger model of about 650 million parameters, SwinV2-H, it achieves 87.1% top-1 accuracy on ImageNet-1K using only ImageNet-1K data. We also leverage this approach to facilitate the training of a 3B model (SwinV2-G), that by $40\times$ less data than that in previous practice, we achieve the state-of-the-art on four representative vision benchmarks. The code and models will be publicly available at https://github.com/microsoft/SimMIM.