With the success of Vision Transformers (ViTs) in computer vision tasks, recent arts try to optimize the performance and complexity of ViTs to enable efficient deployment on mobile devices. Multiple approaches are proposed to accelerate attention mechanism, improve inefficient designs, or incorporate mobile-friendly lightweight convolutions to form hybrid architectures. However, ViT and its variants still have higher latency or considerably more parameters than lightweight CNNs, even true for the years-old MobileNet. In practice, latency and size are both crucial for efficient deployment on resource-constraint hardware. In this work, we investigate a central question, can transformer models run as fast as MobileNet and maintain a similar size? We revisit the design choices of ViTs and propose a novel supernet with low latency and high parameter efficiency. We further introduce a novel fine-grained joint search strategy for transformer models that can find efficient architectures by optimizing latency and number of parameters simultaneously. The proposed models, EfficientFormerV2, achieve 3.5% higher top-1 accuracy than MobileNetV2 on ImageNet-1K with similar latency and parameters. This work demonstrate that properly designed and optimized vision transformers can achieve high performance even with MobileNet-level size and speed.
DETR is a recently proposed Transformer-based method which views object detection as a set prediction problem and achieves state-of-the-art performance but demands extra-long training time to converge. In this paper, we investigate the causes of the optimization difficulty in the training of DETR. Our examinations reveal several factors contributing to the slow convergence of DETR, primarily the issues with the Hungarian loss and the Transformer cross-attention mechanism. To overcome these issues we propose two solutions, namely, TSP-FCOS (Transformer-based Set Prediction with FCOS) and TSP-RC…
Convolutional Neural Networks (ConvNets) are commonly developed at a fixed resource budget, and then scaled up for better accuracy if more resources are available. In this paper, we systematically study model scaling and identify that carefully balancing network depth, width, and resolution can lead to better performance. Based on this observation, we propose a new scaling method that uniformly scales all dimensions of depth/width/resolution using a simple yet highly effective compound coefficient. We demonstrate the effectiveness of this method on scaling up MobileNets and ResNet. To go even …
The dot product self-attention is known to be central and indispensable to state-of-the-art Transformer models. But is it really required? This paper investigates the true importance and contribution of the dot product-based self-attention mechanism on the performance of Transformer models. Via extensive experiments, we find that (1) random alignment matrices surprisingly perform quite competitively and (2) learning attention weights from token-token (query-key) interactions is useful but not that important after all. To this end, we propose \textsc{Synthesizer}, a model that learns synthetic …
Northeastern University（东北大学）；Key Laboratory of Intelligent Computing in Medical Image of Ministry of Education, Northeastern University（东北大学医学影像智能计算教育部重点实验室）；Amii, University of Alberta（阿尔伯塔大学阿尔伯塔机器智能研究所）
Most recent semantic segmentation methods adopt a U-Net framework with an encoder-decoder architecture. It is still challenging for U-Net with a simple skip connection scheme to model the global multi-scale context: 1) Not each skip connection setting is effective due to the issue of incompatible feature sets of encoder and decoder stage, even some skip connection negatively influence the segmentation performance; 2) The original U-Net is worse than the one without any skip connection on some datasets. Based on our findings, we propose a new segmentation framework, named UCTransNet (with a pro…
Decentralized Attention Fails Centralized Signals: Rethinking Transformers for Medical Time Series
Department of Biomedical Engineering, PolyU（PolyU 生物医学工程系）；School of Nursing, PolyU（PolyU 护理学院）；Yau Mathematical Sciences Center, Tsinghua University（清华大学尤太数学科学中心）
NN-Former: Rethinking Graph Structure in Neural Architecture Representation
State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University（多媒体信息处理国家重点实验室，计算机学院，北京大学）；Northwestern Polytechnical University（西北工业大学）；Harbin Institute of Technology (Shenzhen)（哈尔滨工业大学（深圳））
The standard normalization method for neural network (NN) models used in Natural Language Processing (NLP) is layer normalization (LN). This is different than batch normalization (BN), which is widely-adopted in Computer Vision. The preferred use of LN in NLP is principally due to the empirical observation that a (naive/vanilla) use of BN leads to significant performance degradation for NLP tasks; however, a thorough understanding of the underlying reasons for this is not always evident. In this paper, we perform a systematic study of NLP transformer models to understand why BN has a poor perf…
Hong Kong University of Science and Technology（香港科技大学）；The University of Hong Kong（香港大学）；Huawei Noah’s Ark Lab（华为诺亚方舟实验室）；Sun Yat-sen University（中山大学）
Transformer-based models are popularly used in natural language processing (NLP). Its core component, self-attention, has aroused widespread interest. To understand the self-attention mechanism, a direct method is to visualize the attention map of a pre-trained model. Based on the patterns observed, a series of efficient Transformers with different sparse attention masks have been proposed. From a theoretical perspective, universal approximability of Transformer-based models is also recently proved. However, the above understanding and analysis of self-attention is based on a pre-trained model…
We introduce blueprint separable convolutions (BSConv) as highly efficient building blocks for CNNs. They are motivated by quantitative analyses of kernel properties from trained models, which show the dominance of correlations along the depth axis. Based on our findings, we formulate a theoretical foundation from which we derive efficient implementations using only standard layers. Moreover, our approach provides a thorough theoretical derivation, interpretation, and justification for the application of depthwise separable convolutions (DSCs) in general, which have become the basis of many mo…
Google（谷歌）；University of Cambridge（剑桥大学）；Alan Turing Institute（艾伦·图灵研究所）；DeepMind（深度思维）
We introduce Performers, Transformer architectures which can estimate regular (softmax) full-rank-attention Transformers with provable accuracy, but using only linear (as opposed to quadratic) space and time complexity, without relying on any priors such as sparsity or low-rankness. To approximate softmax attention-kernels, Performers use a novel Fast Attention Via positive Orthogonal Random features approach (FAVOR+), which may be of independent interest for scalable kernel methods. FAVOR+ can be also used to efficiently model kernelizable attention mechanisms beyond softmax. This representat…
Microsoft（微软）；University of Science and Technology of China（中国科学技术大学）
We present Mobile-Former, a parallel design of MobileNet and transformer with a two-way bridge in between. This structure leverages the advantages of MobileNet at local processing and transformer at global interaction. And the bridge enables bidirectional fusion of local and global features. Different from recent works on vision transformer, the transformer in Mobile-Former contains very few tokens (e.g. 6 or fewer tokens) that are randomly initialized to learn global priors, resulting in low computational cost. Combining with the proposed light-weight cross attention to model the bridge, Mobi…
Ecole Nationale Supérieure d’Informatique（国立高等计算机学院）；Université Polytechnique Hauts-de-France（上法兰西理工大学）
Vision Transformers have enabled recent attention-based Deep Learning (DL) architectures to achieve remarkable results in Computer Vision (CV) tasks. However, due to the extensive computational resources required, these architectures are rarely implemented on resource-constrained platforms. Current research investigates hybrid handcrafted convolution-based and attention-based models for CV tasks such as image classification and object detection. In this paper, we propose HyT-NAS, an efficient Hardware-aware Neural Architecture Search (HW-NAS) including hybrid architectures targeting vision tas…
Attention-based models have been widely used in many areas, such as computer vision and natural language processing. However, relevant applications in time series classification (TSC) have not been explored deeply yet, causing a significant number of TSC algorithms still suffer from general problems of attention mechanism, like quadratic complexity. In this paper, we promote the efficiency and performance of the attention mechanism by proposing our flexible multi-head linear attention (FMLA), which enhances locality awareness by layer-wise interactions with deformable convolutional blocks and …
Rethinking Nonlinearity: Trainable Gaussian Mixture Modules for Modern Neural Architectures
Guangdong Institute of Intelligence Science and Technology（广东智能科学与技术研究院）；Great Bay University（大湾大学）；University of Massachusetts at Amherst（马萨诸塞大学阿姆赫斯特分校）
Huazhong University of Science & Technology（华中科技大学）；Horizon Robotics（地平线机器人）
Can Transformer perform 2D object- and region-level recognition from a pure sequence-to-sequence perspective with minimal knowledge about the 2D spatial structure? To answer this question, we present You Only Look at One Sequence (YOLOS), a series of object detection models based on the vanilla Vision Transformer with the fewest possible modifications, region priors, as well as inductive biases of the target task. We find that YOLOS pre-trained on the mid-sized ImageNet-1k dataset only can already achieve quite competitive performance on the challenging COCO object detection benchmark, e.g., Y…
Vision Transformers (ViT) have shown rapid progress in computer vision tasks, achieving promising results on various benchmarks. However, due to the massive number of parameters and model design, \textit{e.g.}, attention mechanism, ViT-based models are generally times slower than lightweight convolutional networks. Therefore, the deployment of ViT for real-time applications is particularly challenging, especially on resource-constrained hardware such as mobile devices. Recent efforts try to reduce the computation complexity of ViT through network architecture search or hybrid design with Mobil…
nnMobileNet: Rethinking CNN for Retinopathy Research
School of Computing and Augmented Intelligence, Arizona State University（亚利桑那州立大学计算与增强智能学院）；School of Computing, Clemson University（克莱姆森大学计算学院）；Children’s Hospital Los Angeles（洛杉矶儿童医院）；Mayo Clinic（梅奥诊所）
本文针对 ViT 在视网膜疾病诊断中数据与计算需求高、小病灶定位困难的问题，选择性改进 MobileNet，提出 nnMobileNet，并在多个眼底基准上超过 ViT 模型。
Very recently, Window-based Transformers, which computed self-attention within non-overlapping local windows, demonstrated promising results on image classification, semantic segmentation, and object detection. However, less study has been devoted to the cross-window connection which is the key element to improve the representation ability. In this work, we revisit the spatial shuffle as an efficient way to build connections among windows. As a result, we propose a new vision transformer, named Shuffle Transformer, which is highly efficient and easy to implement by modifying two lines of code.…
Tsinghua Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生院，清华大学）；ByteDance Inc.（字节跳动公司）
The vanilla self-attention mechanism inherently relies on pre-defined and steadfast computational dimensions. Such inflexibility restricts it from possessing context-oriented generalization that can bring more contextual cues and global representations. To mitigate this issue, we propose a Scalable Self-Attention (SSA) mechanism that leverages two scaling factors to release dimensions of query, key, and value matrices while unbinding them with the input. This scalability fetches context-oriented generalization and enhances object sensitivity, which pushes the whole network into a more effectiv…
Light-weight convolutional neural networks (CNNs) are the de-facto for mobile vision tasks. Their spatial inductive biases allow them to learn representations with fewer parameters across different vision tasks. However, these networks are spatially local. To learn global representations, self-attention-based vision trans-formers (ViTs) have been adopted. Unlike CNNs, ViTs are heavy-weight. In this paper, we ask the following question: is it possible to combine the strengths of CNNs and ViTs to build a light-weight and low latency network for mobile vision tasks? Towards this end, we introduce…
Chongqing University of Technology（重庆理工大学）；University of California（加利福尼亚大学）
Recent advances of Transformers have brought new trust to computer vision tasks. However, on small dataset, Transformers is hard to train and has lower performance than convolutional neural networks. We make vision transformers as data-efficient as convolutional neural networks by introducing multi-focal attention bias. Inspired by the attention distance in a well-trained ViT, we constrain the self-attention of ViT to have multi-scale localized receptive field. The size of receptive field is adaptable during training so that optimal configuration can be learned. We provide empirical evidence t…
Huazhong University of Science and Technology（华中科技大学）；Hong Kong University of Science and Technology（香港科技大学）
Vision transformers have recently set off a new wave in the field of medical image analysis due to their remarkable performance on various computer vision tasks. However, recent hybrid-/transformer-based approaches mainly focus on the benefits of transformers in capturing long-range dependency while ignoring the issues of their daunting computational complexity, high training costs, and redundant dependency. In this paper, we propose to employ adaptive pruning to transformers for medical image segmentation and propose a lightweight and effective hybrid network APFormer. To our best knowledge, …
Vision Transformer (ViT) extends the application range of transformers from language processing to computer vision tasks as being an alternative architecture against the existing convolutional neural networks (CNN). Since the transformer-based architecture has been innovative for computer vision modeling, the design convention towards an effective architecture has been less studied yet. From the successful design principles of CNN, we investigate the role of spatial dimension conversion and its effectiveness on transformer-based architecture. We particularly attend to the dimension reduction p…
Technical University of Munich（慕尼黑工业大学）；East China Normal University（华东师范大学）
With the development of the self-attention mechanism, the Transformer model has demonstrated its outstanding performance in the computer vision domain. However, the massive computation brought from the full attention mechanism became a heavy burden for memory consumption. Sequentially, the limitation of memory reduces the possibility of improving the Transformer model. To remedy this problem, we propose a novel memory economy attention mechanism named Couplformer, which decouples the attention map into two sub-matrices and generates the alignment scores from spatial information. A series of di…
Shanghai Jiao Tong University（上海交通大学）；South China University of Technology（南方科技大学）；Beihang University（北航）；Huawei Technologies Ltd（华为技术有限公司）；Shanghai AI Lab（上海人工智能实验室）；Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））
Aerospace Information Research Institute, Chinese Academy of Sciences（中国科学院空天信息创新研究院）；University of Chinese Academy of Sciences（中国科学院大学）；University of Extremadura（埃斯特雷马杜拉大学）；Univ. Grenoble Alpes（格勒诺布尔阿尔卑斯大学）；INRIA（法国国家信息与自动化研究所）；CNRS（法国国家科学研究中心）；Grenoble INP（格勒诺布尔理工学院）
Hyperspectral (HS) images are characterized by approximately contiguous spectral information, enabling the fine identification of materials by capturing subtle spectral discrepancies. Owing to their excellent locally contextual modeling ability, convolutional neural networks (CNNs) have been proven to be a powerful feature extractor in HS image classification. However, CNNs fail to mine and represent the sequence attributes of spectral signatures well due to the limitations of their inherent network backbone. To solve this issue, we rethink HS image classification from a sequential perspective…
Vision Transformers have achieved state-of-the-art performance in many visual tasks. Due to the quadratic computational and memory complexities of self-attention, recent works either apply attention only to low-resolution inputs or restrict the receptive field to a small local region. To overcome these limitations, we propose key-only attention, which excludes query-key pairwise interactions and uses a compute-efficient saliency-gate to obtain attention weights, modeling local-global interactions in all stages. Key-only attention has linear computational and memory complexities w.r.t input siz…
Skip connection, is a widely-used technique to improve the performance and the convergence of deep neural networks, which is believed to relieve the difficulty in optimization due to non-linearity by propagating a linear component through the neural network layers. However, from another point of view, it can also be seen as a modulating mechanism between the input and the output, with the input scaled by a pre-defined value one. In this work, we investigate how the scale factors in the effectiveness of the skip connection and reveal that a trivial adjustment of the scale will lead to spurious …
Exploring sample relationships within each mini-batch has shown great potential for learning image representations. Existing works generally adopt the regular Transformer to model the visual content relationships, ignoring the cues of semantic/label correlations between samples. Also, they generally adopt the "full" self-attention mechanism which are obviously redundant and also sensitive to the noisy samples. To overcome these issues, in this paper, we design a simple yet flexible Batch-Graph Transformer (BGFormer) for mini-batch sample representations by deeply capturing the relationships of…
Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生院）；Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences（中国科学院深圳先进技术研究院）；Shanghai AI Laboratory（上海人工智能实验室）；The University of Sydney（悉尼大学）；University of Chinese Academy of Sciences（中国科学院大学）；Tencent PCG（腾讯平台与内容事业群）
The alignment of adjacent frames is considered an essential operation in video super-resolution (VSR). Advanced VSR models, including the latest VSR Transformers, are generally equipped with well-designed alignment modules. However, the progress of the self-attention mechanism may violate this common sense. In this paper, we rethink the role of alignment in VSR Transformers and make several counter-intuitive observations. Our experiments show that: (i) VSR Transformers can directly utilize multi-frame information from unaligned videos, and (ii) existing alignment methods are sometimes harmful …
Sun Yat-sen University（中山大学）；Microsoft Research Asia（微软亚洲研究院）
Relative position encoding (RPE) is important for transformer to capture sequence ordering of input tokens. General efficacy has been proven in natural language processing. However, in computer vision, its efficacy is not well studied and even remains controversial, e.g., whether relative position encoding can work equally well as absolute position? In order to clarify this, we first review existing relative position encoding methods and analyze their pros and cons when applied in vision transformers. We then propose new relative position encoding methods dedicated to 2D images, called image R…
Rethinking Local Perception in Lightweight Vision Transformer
Tsinghua University（清华大学）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）
In this paper, we study the problem of text line recognition. Unlike most approaches targeting specific domains such as scene-text or handwritten documents, we investigate the general problem of developing a universal architecture that can extract text from any image, regardless of source or input modality. We consider two decoder families (Connectionist Temporal Classification and Transformer) and three encoder modules (Bidirectional LSTMs, Self-Attention, and GRCLs), and conduct extensive experiments to compare their accuracy and performance on widely used public datasets of scene and handwr…
Fudan University（复旦大学）；University of Oxford（牛津大学）；University of Surrey（萨里大学）；Tencent Youtu Lab（腾讯优图实验室）；Facebook AI（脸书人工智能）
Most recent semantic segmentation methods adopt a fully-convolutional network (FCN) with an encoder-decoder architecture. The encoder progressively reduces the spatial resolution and learns more abstract/semantic visual concepts with larger receptive fields. Since context modeling is critical for segmentation, the latest efforts have been focused on increasing the receptive field, through either dilated/atrous convolutions or inserting attention modules. However, the encoder-decoder based FCN architecture remains unchanged. In this paper, we aim to provide an alternative perspective by treatin…
JD Explore Academy（京东探索研究院）；The University of Sydney（悉尼大学）
Self-supervised pre-training vision transformer (ViT) via masked image modeling (MIM) has been proven very effective. However, customized algorithms should be carefully designed for the hierarchical ViTs, e.g., GreenMIM, instead of using the vanilla and simple MAE for the plain ViT. More importantly, since these hierarchical ViTs cannot reuse the off-the-shelf pre-trained weights of the plain ViTs, the requirement of pre-training them leads to a massive amount of computational cost, thereby incurring both algorithmic and computational complexity. In this paper, we address this problem by propo…
Rethinking Attention Gated with Hybrid Dual Pyramid Transformer-CNN for Generalized Segmentation in Medical Imaging
Junia（朱尼亚学院）；Centrale Lille（里尔中央理工学院）；University of the Basque Country UPV/EHU（巴斯克大学 UPV/EHU）；Université Polytechnique Hauts-de-France（上法兰西理工大学）；Université de Lille（里尔大学）；Ho Chi Minh City Open University（胡志明市开放大学）
