This paper presents a new vision Transformer, called Swin Transformer, that capably serves as a general-purpose backbone for computer vision. Challenges in adapting Transformer from language to vision arise from differences between the two domains, such as large variations in the scale of visual entities and the high resolution of pixels in images compared to words in text. To address these differences, we propose a hierarchical Transformer whose representation is computed with shifted windows. The shifted windowing scheme brings greater efficiency by limiting self-attention computation to non-overlapping local windows while also allowing for cross-window connection. This hierarchical architecture has the flexibility to model at various scales and has linear computational complexity with respect to image size. These qualities of Swin Transformer make it compatible with a broad range of vision tasks, including image classification (86.4 top-1 accuracy on ImageNet-1K) and dense prediction tasks such as object detection (58.7 box AP and 51.1 mask AP on COCO test-dev) and semantic segmentation (53.5 mIoU on ADE20K val). Its performance surpasses the previous state-of-the-art by a large margin of +2.7 box AP and +2.6 mask AP on COCO, and +3.2 mIoU on ADE20K, demonstrating the potential of Transformer-based models as vision backbones. The code and models will be made publicly available at~\url{this https URL}.

Swin Transformer: Hierarchical Vision Transformer using Shifted Windows.

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 treating semantic segmentation as a sequence-to-sequence prediction task. Specifically, we deploy a pure transformer (i.e., without convolution and resolution reduction) to encode an image as a sequence of patches. With the global context modeled in every layer of the transformer, this encoder can be combined with a simple decoder to provide a powerful segmentation model, termed SEgmentation TRansformer (SETR). Extensive experiments show that SETR achieves new state of the art on ADE20K (50.28% mIoU), Pascal Context (55.83% mIoU) and competitive results on Cityscapes. Particularly, we achieve the first position in the highly competitive ADE20K test server leaderboard on the day of submission.

/pdf/rethinking-semantic-segmentation-from-a-sequence-to-sequence-kn491bce89.pdf

Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers

Recently, channel attention mechanism has demonstrated to offer great potential in improving the performance of deep convolutional neural networks (CNNs). However, most existing methods dedicate to developing more sophisticated attention modules for achieving better performance, which inevitably increase model complexity. To overcome the paradox of performance and complexity trade-off, this paper proposes an Efficient Channel Attention (ECA) module, which only involves a handful of parameters while bringing clear performance gain. By dissecting the channel attention module in SENet, we empirically show avoiding dimensionality reduction is important for learning channel attention, and appropriate cross-channel interaction can preserve performance while significantly decreasing model complexity. Therefore, we propose a local cross-channel interaction strategy without dimensionality reduction, which can be efficiently implemented via 1D convolution. Furthermore, we develop a method to adaptively select kernel size of 1D convolution, determining coverage of local cross-channel interaction. The proposed ECA module is both efficient and effective, e.g., the parameters and computations of our modules against backbone of ResNet50 are 80 vs. 24.37M and 4.7e-4 GFlops vs. 3.86 GFlops, respectively, and the performance boost is more than 2% in terms of Top-1 accuracy. We extensively evaluate our ECA module on image classification, object detection and instance segmentation with backbones of ResNets and MobileNetV2. The experimental results show our module is more efficient while performing favorably against its counterparts.

/pdf/eca-net-efficient-channel-attention-for-deep-convolutional-2f1b7euz6m.pdf

ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks

Recent studies on mobile network design have demonstrated the remarkable effectiveness of channel attention (e.g., the Squeeze-and-Excitation attention) for lifting model performance, but they generally neglect the positional information, which is important for generating spatially selective attention maps. In this paper, we propose a novel attention mechanism for mobile networks by embedding positional information into channel attention, which we call "coordinate attention". Unlike channel attention that transforms a feature tensor to a single feature vector via 2D global pooling, the coordinate attention factorizes channel attention into two 1D feature encoding processes that aggregate features along the two spatial directions, respectively. In this way, long-range dependencies can be captured along one spatial direction and meanwhile precise positional information can be preserved along the other spatial direction. The resulting feature maps are then encoded separately into a pair of direction-aware and position-sensitive attention maps that can be complementarily applied to the input feature map to augment the representations of the objects of interest. Our coordinate attention is simple and can be flexibly plugged into classic mobile networks, such as MobileNetV2, MobileNeXt, and EfficientNet with nearly no computational overhead. Extensive experiments demonstrate that our coordinate attention is not only beneficial to ImageNet classification but more interestingly, behaves better in down-stream tasks, such as object detection and semantic segmentation. Code is available at https://github.com/Andrew-Qibin/CoordAttention.

/pdf/coordinate-attention-for-efficient-mobile-network-design-19ipozl9y3.pdf

Coordinate Attention for Efficient Mobile Network Design

Full-image dependencies provide useful contextual information to benefit visual understanding problems. In this work, we propose a Criss-Cross Network (CCNet) for obtaining such contextual information in a more effective and efficient way. Concretely, for each pixel, a novel criss-cross attention module in CCNet harvests the contextual information of all the pixels on its criss-cross path. By taking a further recurrent operation, each pixel can finally capture the full-image dependencies from all pixels. Overall, CCNet is with the following merits: 1) GPU memory friendly. Compared with the non-local block, the proposed recurrent criss-cross attention module requires 11x less GPU memory usage. 2) High computational efficiency. The recurrent criss-cross attention significantly reduces FLOPs by about 85% of the non-local block in computing full-image dependencies. 3) The state-of-the-art performance. We conduct extensive experiments on popular semantic segmentation benchmarks including Cityscapes, ADE20K, and instance segmentation benchmark COCO. In particular, our CCNet achieves the mIoU score of 81.4 and 45.22 on Cityscapes test set and ADE20K validation set, respectively, which are the new state-of-the-art results. The source code is available at https://github.com/speedinghzl/CCNet.

/pdf/ccnet-criss-cross-attention-for-semantic-segmentation-2m8k1ptda1.pdf

CCNet: Criss-Cross Attention for Semantic Segmentation

In this paper, we address the scene segmentation task by capturing rich contextual dependencies based on the self-attention mechanism. Unlike previous works that capture contexts by multi-scale features fusion, we propose a Dual Attention Networks (DANet) to adaptively integrate local features with their global dependencies. Specifically, we append two types of attention modules on top of traditional dilated FCN, which model the semantic interdependencies in spatial and channel dimensions respectively. The position attention module selectively aggregates the features at each position by a weighted sum of the features at all positions. Similar features would be related to each other regardless of their distances. Meanwhile, the channel attention module selectively emphasizes interdependent channel maps by integrating associated features among all channel maps. We sum the outputs of the two attention modules to further improve feature representation which contributes to more precise segmentation results. We achieve new state-of-the-art segmentation performance on three challenging scene segmentation datasets, i.e., Cityscapes, PASCAL Context and COCO Stuff dataset. In particular, a Mean IoU score of 81.5% on Cityscapes test set is achieved without using coarse data.

/pdf/dual-attention-network-for-scene-segmentation-1t75tyo5w8.pdf

Dual Attention Network for Scene Segmentation

In this article, we propose a Dual Relation-aware Attention Network (DRANet) to handle the task of scene segmentation. How to efficiently exploit context is essential for pixel-level recognition. To address the issue, we adaptively capture contextual information based on the relation-aware attention mechanism. Especially, we append two types of attention modules on the top of the dilated fully convolutional network (FCN), which model the contextual dependencies in spatial and channel dimensions, respectively. In the attention modules, we adopt a self-attention mechanism to model semantic associations between any two pixels or channels. Each pixel or channel can adaptively aggregate context from all pixels or channels according to their correlations. To reduce the high cost of computation and memory caused by the abovementioned pairwise association computation, we further design two types of compact attention modules. In the compact attention modules, each pixel or channel is built into association only with a few numbers of gathering centers and obtains corresponding context aggregation over these gathering centers. Meanwhile, we add a cross-level gating decoder to selectively enhance spatial details that boost the performance of the network. We conduct extensive experiments to validate the effectiveness of our network and achieve new state-of-the-art segmentation performance on four challenging scene segmentation data sets, i.e., Cityscapes, ADE20K, PASCAL Context, and COCO Stuff data sets. In particular, a Mean IoU score of 82.9% on the Cityscapes test set is achieved without using extra coarse annotated data.

Scene Segmentation With Dual Relation-Aware Attention Network

Recent works attempt to improve scene parsing performance by exploring different levels of contexts, and typically train a well-designed convolutional network to exploit useful contexts across all pixels equally. However, in this paper, we find that the context demands are varying from different pixels or regions in each image. Based on this observation, we propose an Adaptive Context Network (ACNet) to capture the pixel-aware contexts by a competitive fusion of global context and local context according to different per-pixel demands. Specifically, when given a pixel, the global context demand is measured by the similarity between the global feature and its local feature, whose reverse value can also be used to measure the local context demand. We model the two demanding measurements by the proposed global context module and local context module, respectively, to generate their adaptive contextual features. Furthermore, we import multiple such modules to build several adaptive context blocks in different levels of network to obtain a coarse-to-fine result. Finally, comprehensive experimental evaluations demonstrate the effectiveness of the proposed ACNet, and new state-of-the-arts performances are achieved on all four public datasets, i.e. Cityscapes, ADE20K, PASCAL Context, and COCO Stuff.

/pdf/adaptive-context-network-for-scene-parsing-2mmjaotpp8.pdf

Adaptive Context Network for Scene Parsing

In this paper, we address the scene segmentation task by capturing rich contextual dependencies based on the selfattention mechanism. Unlike previous works that capture contexts by multi-scale features fusion, we propose a Dual Attention Networks (DANet) to adaptively integrate local features with their global dependencies. Specifically, we append two types of attention modules on top of traditional dilated FCN, which model the semantic interdependencies in spatial and channel dimensions respectively. The position attention module selectively aggregates the features at each position by a weighted sum of the features at all positions. Similar features would be related to each other regardless of their distances. Meanwhile, the channel attention module selectively emphasizes interdependent channel maps by integrating associated features among all channel maps. We sum the outputs of the two attention modules to further improve feature representation which contributes to more precise segmentation results. We achieve new state-of-the-art segmentation performance on three challenging scene segmentation datasets, i.e., Cityscapes, PASCAL Context and COCO Stuff dataset. In particular, a Mean IoU score of 81.5% on Cityscapes test set is achieved without using coarse data. We make the code and trained model publicly available at this https URL

Recent works attempt to improve scene parsing performance by exploring different levels of contexts, and typically train a well-designed convolutional network to exploit useful contexts across all pixels equally. However, in this paper, we find that the context demands are varying from different pixels or regions in each image. Based on this observation, we propose an Adaptive Context Network (ACNet) to capture the pixel-aware contexts by a competitive fusion of global context and local context according to different per-pixel demands. Specifically, when given a pixel, the global context demand is measured by the similarity between the global feature and its local feature, whose reverse value can be used to measure the local context demand. We model the two demand measurements by the proposed global context module and local context module, respectively, to generate adaptive contextual features. Furthermore, we import multiple such modules to build several adaptive context blocks in different levels of network to obtain a coarse-to-fine result. Finally, comprehensive experimental evaluations demonstrate the effectiveness of the proposed ACNet, and new state-of-the-arts performances are achieved on all four public datasets, i.e. Cityscapes, ADE20K, PASCAL Context, and COCO Stuff.

Yong Li

Papers

Dual Attention Network for Scene Segmentation

Scene Segmentation With Dual Relation-Aware Attention Network

Adaptive Context Network for Scene Parsing

Dual Attention Network for Scene Segmentation

Adaptive Context Network for Scene Parsing