paper-with-me

홈 › Papers

Dual-former: Hybrid Self-attention Transformer for Efficient Image Restoration

2022-10-03 · Sixiang Chen, Tian Ye, Yun Liu, ErKang Chen

Recently, image restoration transformers have achieved comparable performance with previous state-of-the-art CNNs. However, how to efficiently leverage such architectures remains an open problem. In this work, we present Dual-former whose critical insight is to combine the powerful global modeling ability of self-attention modules and the local modeling ability of convolutions in an overall architecture. With convolution-based Local Feature Extraction modules equipped in the encoder and the decoder, we only adopt a novel Hybrid Transformer Block in the latent layer to model the long-distance dependence in spatial dimensions and handle the uneven distribution between channels. Such a design eliminates the substantial computational complexity in previous image restoration transformers and achieves superior performance on multiple image restoration tasks. Experiments demonstrate that Dual-former achieves a 1.91dB gain over the state-of-the-art MAXIM method on the Indoor dataset for single image dehazing while consuming only 4.2% GFLOPs as MAXIM. For single image deraining, it exceeds the SOTA method by 0.1dB PSNR on the average results of five datasets with only 21.5% GFLOPs. Dual-former also substantially surpasses the latest desnowing method on various datasets, with fewer parameters.

📄 PDF Abstract BibTeX arXiv:2210.01069

Code (0)

등록된 구현이 없습니다.

Tasks

DecoderImage DehazingImage RestorationRain RemovalSingle Image DehazingSingle Image Deraining

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Adam 설명 없음
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Position-Wise Feed-Forward Layer 설명 없음

Similar Papers 제목 키워드 기반

HAAT: Hybrid Attention Aggregation Transformer for Image Super-Resolution

2024-11-27 · Song-Jiang Lai, Tsun-Hin Cheung, Ka-Chun Fung, Kai-wen Xue 외

In the research area of image super-resolution, Swin-transformer-based models are favored for their global spatial modeling and shifting window attention mechanism. However, existing methods often limit self-attention to…

Image Super-ResolutionSuper-Resolution

A survey of the Vision Transformers and their CNN-Transformer based Variants

2023-05-17 · Asifullah Khan, Zunaira Rauf, Anabia Sohail, Abdul Rehman 외

Vision transformers have become popular as a possible substitute to convolutional neural networks (CNNs) for a variety of computer vision applications. These transformers, with their ability to focus on global relationsh…

Survey

Accelerating Transformer Decoding via a Hybrid of Self-attention and Recurrent Neural Network

2019-09-05 · Chengyi Wang, Shuangzhi Wu, Shujie Liu

Due to the highly parallelizable architecture, Transformer is faster to train than RNN-based models and popularly used in machine translation tasks. However, at inference time, each output word requires all the hidden st…

DecoderKnowledge DistillationMachine TranslationTranslation

DIAMANT: Dual Image-Attention Map Encoders For Medical Image Segmentation

2023-04-28 · Yousef Yeganeh, Azade Farshad, Peter Weinberger, Seyed-Ahmad Ahmadi 외

Although purely transformer-based architectures showed promising performance in many computer vision tasks, many hybrid models consisting of CNN and transformer blocks are introduced to fit more specialized tasks. Nevert…

Image SegmentationMedical Image SegmentationSemantic Segmentation

Integrating Quantum-Classical Attention in Patch Transformers for Enhanced Time Series Forecasting

2025-03-31 · Sanjay Chakraborty, Fredrik Heintz

QCAAPatchTF is a quantum attention network integrated with an advanced patch-based transformer, designed for multivariate time series forecasting, classification, and anomaly detection. Leveraging quantum superpositions,…

Anomaly DetectionMultivariate Time Series ForecastingTime SeriesTime Series Forecasting