Search

To pursue high accuracy, most image semantic segmentation methods are computationally costly and thus not suitable to real-time applications. Existing lightweight methods either adopt a single branch without feature fusion, which damages accuracy, or introduce extra branches for feature fusion, which harms efficiency. In this paper, we propose a lightweight network named EFRNet, with feature fusion and refinement in a single branch to achieve better balance between accuracy and efficiency in real-time semantic segmentation. Specifically, in EFRNet, we design a novel Feature Fusion Module to fuse multi-stage features in a single CNN efficiently, and we propose a lightweight Channel Attention Refinement Module to refine features with few extra parameters. Extensive experiments show that our EFRNet achieves decent accuracy with an extremely small model size and high inference speed. It achieves the best accuracy of 70.02% mIoU compared with state-of-the-art lightweight methods on CamVid with only 0.48M parameters. © 2021 IEEE

For most of the non-Gaussian statistical models, the data being modeled represent strongly structured properties, such as scalar data with bounded support (e.g., beta distribution), vector data with unit length (e.g., Dirichlet distribution), and vector data with positive elements (e.g., generalized inverted Dirichlet distribution). In practical implementations of non-Gaussian statistical models, it is infeasible to find an analytically tractable solution to estimating the posterior distributions of the parameters. Variational inference (VI) is a widely used framework in Bayesian estimation. Recently, an improved framework, namely, the extended VI (EVI), has been introduced and applied successfully to a number of non-Gaussian statistical models. EVI derives analytically tractable solutions by introducing lower bound approximations to the variational objective function. In this paper, we compare two approximation strategies, namely, the multiple lower bounds (MLBs) approximation and the single lower bound (SLB) approximation, which can be applied to carry out the EVI. For implementation, two different conditions, the weak and the strong conditions, are discussed. Convergence of the EVI depends on the selection of the lower bound, regardless of the choice of weak or strong condition. We also discuss the convergence properties to clarify the differences between MLB and SLB. Extensive comparisons are made based on some EVI-based non-Gaussian statistical models. Theoretical analysis is conducted to demonstrate the differences between the weak and strong conditions. Experimental results based on real data show advantages of the SLB approximation over the MLB approximation.

Defocus map estimation (DME) is very useful in many computer vision applications and has drawn much attention in recent years. Edge-based DME methods can generate sharp defocus discontinuities but usually suffer from textures of the input image. Region-based methods are free of textures but cannot catch the defocus discontinuities very well. In this paper, we propose a DME method combining edge-based and region-based methods together to keep their respective advantages while eliminating the shortcomings. The combination is achieved via regression tree fields (RTF). In an RTF, the input feature and the linear basis are of vital importance. For our RTF, they are obtained as follows. (i) Two orthogonal gradient operators with the corresponding subsets of Gabor filters are employed in localized 2D frequency analysis to generate accurate likelihood, and the first K highest local maximums of likelihood are sent to an RTF as input feature. (ii) At the same time, the input image is processed by three edge-based methods and the results serve as the linear basis of RTF. The experiments demonstrate that the proposed method outperforms state-of-the-art DME methods. Moreover, the proposed method can be readily applied to defocused image deblurring and defocus blur detection.