Search

Image smoothing is a prerequisite for many computer vision and graphics applications. In this article, we raise an intriguing question whether a dataset that semantically describes meaningful structures and unimportant details can facilitate a deep learning model to smooth complex natural images. To answer it, we generate ground-truth labels from easy samples by candidate generation and a screening test and synthesize hard samples in structure-preserving smoothing by blending intricate and multifarious details with the labels. To take full advantage of this dataset, we present a joint edge detection and structure-preserving image smoothing neural network (JESS-Net). Moreover, we propose the distinctive total variation loss as prior knowledge to narrow the gap between synthetic and real data. Experiments on different datasets and real images show clear improvements of our method over the state of the arts in terms of both the image cleanness and structure-preserving ability. Code and dataset are available at https://github.com/YidFeng/Easy2Hard.

This paper looks at this intriguing question: are single images with their details lost during deraining, reversible to their artifact-free status? We propose an end-to-end detail-recovery image deraining network (termed a DRDNet) to solve the problem. Unlike existing image deraining approaches that attempt to meet the conflicting goal of simultaneously deraining and preserving details in a unified framework, we propose to view rain removal and detail recovery as two seperate tasks, so that each part could specialize rather than trade-off between two conflicting goals. Specifically, we introduce two parallel sub-networks with a comprehensive loss function which synergize to derain and recover the lost details caused by deraining. For complete rain removal, we present a rain residual network with the squeeze-and-excitation (SE) operation to remove rain streaks from the rainy images. For detail recovery, we construct a specialized detail repair network consisting of welldesigned blocks, named structure detail context aggregation block (SDCAB), to encourage the lost details to return for eliminating image degradations. Moreover, the detail recovery branch of our proposed detail repair framework is detachable and can be incorporated into existing deraining methods to boost their performances. DRD-Net has been validated on several well-known benchmark datasets in terms of deraining robustness and detail accuracy. Comparisons show clear visual and numerical improvements of our method over the state-of-the-arts.

In this paper, we propose a normal estimation method for unstructured point cloud. We observe that geometric estimators commonly focus more on feature preservation but are hard to tune parameters and sensitive to noise, while learning-based approaches pursue an overall normal estimation accuracy but cannot well handle challenging regions such as surface edges. This paper presents a novel normal estimation method, under the co-support of geometric estimator and deep learning. To lowering the learning difficulty, we first propose to compute a suboptimal initial normal at each point by searching for a best fitting patch. Based on the computed normal field, we design a normal-based height map network (NH-Net) to fine-tune the suboptimal normals. Qualitative and quantitative evaluations demonstrate the clear improvements of our results over both traditional methods and learning-based methods, in terms of estimation accuracy and feature recovery.

Existing texture filtering techniques acquiesce in the removal of all undesired background texture patterns simultaneously, and lead to the disappearance of small-scale foreground structures. Differently, this paper considers both multi-scale structures and specified texture patterns within an image as significant image contents, and presents a content-aware texture filtering approach via an easy-to-use interactive interface, named intuitive single click. The user’s effort on content-aware texture filtering is reduced to only locating a single click, where it is intuitively considered as one type of texture patterns. This simple interface is made possible by first generating three-scale intermediate filtered images (i.e., high, medium and low) and then performing breadth-first search (BFS) to identify one or more texture regions. Specifically, we utilize BFS to purify image edges in the medium-scale filtered image, on top of which a mask representing texture and non-texture regions is abstracted through the intuitive single-click texture locating and BFS. The mask is applied to effectively maintain multi-scale structures in the low-scale filtered image, and guides to fully smooth textures in the high-scale filtered image. Finally, a clean image is yielded by simply synthesizing the results of two mask operations in the low- and high-scale filtered images. We illustrate by extensive experiments that our approach advances the state-of-the-arts in terms of both the image cleanness and content-preserving ability. Our code is publicly available.

Normal filtering is a fundamental step of feature-preserving mesh denoising. Methods based on convolutional neural networks (CNNs) have recently made their debut for normal filtering. However, they require complicated voxelization and/or projection operations for regularization, and afford an overall denoising accuracy with few powers of preserving surface features. We devise a novel normal filtering neural network algorithm, which we call as NormalF-Net. NormalF-Net consists of two cascaded subnetworks with a comprehensive loss function. The first subnetwork learns mapping from non-local patch-group normal matrices (NPNMs) to their ground-truth low-rank counterparts for denoising, and the second subnetwork learns mapping from the recovered NPNMs to the ground-truth normals for normal refinement. Different from existing learning-based methods, NormalF-Net, which bridges the connection between CNNs and geometry domain knowledge of non-local similarity, can not only preserve surface features when removing different levels and types of noise, but be free of voxelization/projection. NormalF-Net has been validated on different datasets of meshes with multi-scale features yet corrupted by noise of different distributions. Experimental results consistently demonstrate clear improvements of our method over the state-of-the-arts in both noise-robustness and feature awareness.

We present a novel direction-aware feature-level frequency decomposition network for single image deraining. Compared with existing solutions, the proposed network has three compelling characteristics. First, unlike previous algorithms, we propose to perform frequency decomposition at feature-level instead of image-level, allowing both low-frequency maps containing structures and high-frequency maps containing details to be continuously refined during the training procedure. Second, we further establish communication channels between low-frequency maps and high-frequency maps to interactively capture structures from high-frequency maps and add them back to low-frequency maps and, simultaneously, extract details from low-frequency maps and send them back to high-frequency maps, thereby removing rain streaks while preserving more delicate features in the input image. Third, different from existing algorithms using convolutional filters consistent in all directions, we propose a direction-aware filter to capture the direction of rain streaks in order to more effectively and thoroughly purge the input images of rain streaks. We extensively evaluate the proposed approach in three representative datasets and experimental results corroborate our approach consistently outperforms state-of-the-art deraining algorithms.