Search

In this paper, we aim to improve super-resolution (SR) imaging quality on 4K/8K images with a negligible increase in computational cost and alleviate the accuracy drop after quantization. Experiments have discovered two phenomena: (1) the re-parameterization (Rep) technique has no apparent advantages in regions with smooth textures, and (2) the accuracy drop after quantization compared with SR methods based on No-Rep because the structure information of the image will be weakened caused by the multi-branch fusion in Rep technique. Inspired by the above phenomenon, we innovatively combine texture classification and Rep techniques to propose a generic TARepSR framework (consisting of TA-Module and VarRepSR-Module) to adjust the branching of Rep blocks texture-awarely. Specifically, the TA-Module is a lightweight classification network to classify textures in different regions. An existing SR network using texture-aware Rep techniques can be used as the VarRepSR-Module to super-resolute images with higher imaging quality without additional computational costs. Moreover, we propose a TC loss to avoid over-fitting caused by an unbalanced degree of the tendency of classification results to classify different textures better. Experiments show that our TARepSR can not only improve the imaging quality of most existing methods (e.g., FSRCNN, CARN, EDSR, XLSR) on 4K/8K images with negligible increase in computational cost but also improve the accuracy after quantization compared with the state-of-the-art Rep methods. [ABSTRACT FROM AUTHOR]

In this paper, we aim to improve the imaging quality of super-resolution (SR) without increasing the inference time to address the difficulty of trading off between quality and inference time in many existing methods and design a deployment-friendly, lightweight model for mobile devices. Specifically, we propose a general RepDFSR framework to enhance the textures of SR images while avoiding additional inference time overhead, which can be applied to existing SR networks. It incorporates innovative convolutional block design, loss function design, and the re-parameterizable technique. In RepDFSR, we propose a re-parameterizable texture-enhanced convolution based on differential filters. It extracts texture information more advantageous and efficient in training than regular convolution. Secondly, we propose a DF loss function to compel the model to super-resolve the gradient mappings with high variance, thus reconstructing images with sharper textures. Moreover, we propose a TELNet network for mobile devices based on the RepDFSR framework to validate the effectiveness of RepDFSR and our thoughts on practical applications of SR on mobile devices. The experimental results demonstrate the successful integration of the RepDFSR framework with the existing SR methods. Additionally, the TELNet meets the requirements of mobile devices for hardware and quantization limitations, showcasing superior SR performance compared to classic and state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Lane detection is a vital component of intelligent driving systems, offering indispensable functionality to keep the vehicle within its designated lane, thereby reducing the risk of lane departure. However, the complexity of the traffic environment, coupled with the rapid movement of vehicles, creates many challenges for detection tasks. Current lane detection methods suffer from issues such as low feature extraction capability, poor real-time detection, and inadequate robustness. Addressing these issues, this paper proposes a lane detection algorithm that combines an online re-parameterization ResNet with a hybrid attention mechanism. Firstly, we replaced standard convolution with online re-parameterization convolution, simplifying the convolutional operations during the inference phase and subsequently reducing the detection time. In an effort to enhance the performance of the model, a hybrid attention module is incorporated to enhance the ability to focus on elongated targets. Finally, a row anchor lane detection method is introduced to analyze the existence and location of lane lines row by row in the image and output the predicted lane positions. The experimental outcomes illustrate that the model achieves F1 scores of 96.84% and 75.60% on the publicly available TuSimple and CULane lane datasets, respectively. Moreover, the inference speed reaches a notable 304 frames per second (FPS). The overall performance outperforms other detection models and fulfills the requirements of real-time responsiveness and robustness for lane detection tasks. [ABSTRACT FROM AUTHOR]

In many speaker verification tasks, the quality of speaker embedding is an important factor in affecting speaker verification systems. Advanced speaker embedding extraction networks aim to capture richer speaker features through the multi-branch network architecture. Recently, speaker verification systems based on transformer encoders have received much attention, and many satisfactory results have been achieved because transformer encoders can efficiently extract the global features of the speaker (e.g., MFA-Conformer). However, the large number of model parameters and computational latency are common problems faced by the above approaches, which make them difficult to apply to resource-constrained edge terminals. To address this issue, this paper proposes an effective, lightweight transformer model (MCA-former) with multi-scale convolutional self-attention (MCA), which can perform multi-scale modeling and channel modeling in the temporal direction of the input with low computational cost. In addition, in the inference phase of the model, we further develop a systematic re-parameterization method to convert the multi-branch network structure into the single-path topology, effectively improving the inference speed. We investigate the performance of the MCA-former for speaker verification under the VoxCeleb1 test set. The results show that the MCA-based transformer model is more advantageous in terms of the number of parameters and inference efficiency. By applying the re-parameterization, the inference speed of the model is increased by about 30%, and the memory consumption is significantly improved. • Designing a lightweight multi-scale convolutional self-attention module. • An efficient transformer encoder for speaker verification. • Using the re-parameterization method improves the model's inference efficiency. [ABSTRACT FROM AUTHOR]

You recklessly told your boss that solving a non-linear system of size n ( n unknowns and n equations) requires a time proportional to n , as you were not very attentive during algorithmic complexity lectures. So now, you have only one night to solve a problem of big size (e.g., 1000 equations/unknowns), otherwise you will be fired in the next morning. The system is well-constrained and structurally irreducible: it does not contain any strictly smaller well-constrained subsystems. Its size is big, so the Newton–Raphson method is too slow and impractical. The most frustrating thing is that if you knew the values of a small number k ≪ n of key unknowns, then the system would be reducible to small square subsystems and easily solved. You wonder if it would be possible to exploit this reducibility, even without knowing the values of these few key unknowns. This article shows that it is indeed possible. This is done at the lowest level, at the linear algebra routines level, so that numerous solvers (Newton–Raphson, homotopy, and also p -adic methods relying on Hensel lifting) widely involved in geometric constraint solving and CAD applications can benefit from this decomposition with minor modifications. For instance, with k ≪ n key unknowns, the cost of a Newton iteration becomes O ( k n 2 ) instead of O ( n 3 ) . Several experiments showing a significant performance gain of our re-parameterization technique are reported in this paper to consolidate our theoretical findings and to motivate its practical usage for bigger systems. [ABSTRACT FROM AUTHOR]

The Ensemble Kalman Filter (EnKF) is a Monte Carlo based method to assimilate the measurement data sequentially in time. Although, EnKF has some advantages over the other Kalman based methods to deal with non-linear and/or high dimensional reservoir models, it also suffers from deficiency in estimation of non-Gaussian parameters. In this work, we propose a re-parameterization method to handle non-Gaussian parameters via Ensemble Kalman Filter framework. For this purpose, concept of cumulative distribution function transformation has been used. In addition, the statistics of prior information have been aggregated in the state vector in order to capture the prior uncertainties of non-Gaussian parameters. To evaluate the performance of the new method, three estimation examples have been implemented. These examples are designed to evaluate the performance of the proposed method to estimate both local and global parameters. The performance of method in handling non-Gaussian prior as well as capturing the prior uncertainty has been assessed. The results revealed that, the proposed method handled both non-Gaussian prior and capturing of the prior uncertainties of the local and global parameters quite efficiently. Compared to previously conducted researches, new algorithm would considerably reduce the number of required ensemble members to converge to the appropriate solution. Standard ensemble Kalman filter and normal score transformation have been also implemented for each setup; the obtained results show that the new proposed method outperforms the other alternate methods. [ABSTRACT FROM AUTHOR]

Highlights: [•] Presents a novel method for both shape design and isogeometric analysis using mapped basis functions. [•] The space spanned by mapped B-spline basis functions is an extension of uniform B-splines over an arbitrary parameterization. [•] The continuity of the resulting surfaces can be arbitrary higher order, including at extraordinary points. [•] The proposed method can be further extended to other basis functions as well as to non-quadrilateral meshes. [ABSTRACT FROM AUTHOR]