Methods In Visual Quality Research Articles

Hierarchical Color Fusion Network (HCFN): Enhancing exemplar-based video colorization

We conducted a study on exemplar-based colorization for grayscale videos, aiming to enhance visual perception by transforming them into vibrant videos with plausible colors. Despite promising results from existing approaches, effectively transferring colors from a reference image to grayscale video frames while maintaining temporal consistency between frames remains a challenge. To tackle this challenge, we designed a Hierarchical Color Fusion Network (HCFN). For each grayscale video frame, HCFN initially uses global and local attention mechanisms to calculate pixel-level similarity with the reference image and spatio-temporal correspondence with its neighboring frame, respectively. Based on these relationships, the pixel-level color and spatio-temporal color of the grayscale video frame are generated. Additionally, the tone-based color for the given frame is obtained based on the color distribution of the reference image. Finally, the ambiguity among these three types of colors is eliminated through the proposed hierarchical fusion mechanism, and the final color of the grayscale video frame is produced. Experimental results on public databases show that our method outperforms state-of-the-art methods in visual quality, realism and temporal consistency by a large margin. The source code and pre-trained checkpoints for HCFN is publicly available at https://github.com/wangyins/HCFN.

FineStyle: Semantic-Aware Fine-Grained Motion Style Transfer with Dual Interactive-Flow Fusion.

We present FineStyle, a novel framework for motion style transfer that generates expressive human animations with specific styles for virtual reality and vision fields. It incorporates semantic awareness, which improves motion representation and allows for precise and stylish animation generation. Existing methods for motion style transfer have all failed to consider the semantic meaning behind the motion, resulting in limited controls over the generated human animations. To improve, FineStyle introduces a new cross-modality fusion module called Dual Interactive-Flow Fusion (DIFF). As the first attempt, DIFF integrates motion style features and semantic flows, producing semantic-aware style codes for fine-grained motion style transfer. FineStyle uses an innovative two-stage semantic guidance approach that leverages semantic clues to enhance the discriminative power of both semantic and style features. At an early stage, a semantic-guided encoder introduces distinct semantic clues into the style flow. Then, at a fine stage, both flows are further fused interactively, selecting the matched and critical clues from both flows. Extensive experiments demonstrate that FineStyle outperforms state-of-the-art methods in visual quality and controllability. By considering the semantic meaning behind motion style patterns, FineStyle allows for more precise control over motion styles. Source code and model are available on https://github.com/XingliangJin/Fine-Style.git.

Underwater image enhancement via two-level wavelet decomposition maximum brightness color restoration and edge refinement histogram stretching.

Underwater images suffer color distortions and low contrast. This is because the light is absorbed and scattered when it travels through water. Different underwater scenes result in different color deviations and levels of detail loss in underwater images. To address these issues of color distortion and low contrast, an underwater image enhancement method that includes two-level wavelet decomposition maximum brightness color restoration, and edge refinement histogram stretching is proposed. First, according to the Jaffe-McGlamery underwater optical imaging model, the proportions of the maximum bright channel were obtained to correct the color of underwater images. Then, edge refinement histogram stretching was designed, and edge refinement and denoising processing were performed while stretching the histogram to enhance contrast and noise removal. Finally, wavelet two-level decomposition of the color-corrected and contrast-stretched underwater images was performed, and the decomposed components in equal proportions were fused. The proposed method can restore the color and detail and enhance the contrast of the underwater image. Extensive experiments demonstrated that the proposed method achieves superior performance against state-of-the-art methods in visual quality and quantitative metrics.

Vision-based aerial image mosaicking algorithm with object detection

Aerial image sequence mosaicking is one of the challenging research fields in computer vision. To obtain large-scale orthophoto maps with object detection information, we propose a vision-based image mosaicking algorithm without any extra location data. According to object detection results, we define a complexity factor to describe the importance of each input Image and dynamically optimize the feature extraction process. The feature points extraction and matching processes are mainly guided by the speeded-up robust features (SURF) and the grid motion statistic (GMS) algorithm respectively. A robust reference frame selection method is proposed to eliminate the transformation distortion by searching for the center area based on overlaps. Besides, the sparse Levenberg-Marquardt (LM) algorithm and the heavy occluded frames removal method are Applied to reduce accumulated errors and further improve the mosaicking performance. The proposed algorithm is performed by using multithreading and graphics processing unit (GPU) acceleration on several aerial image datasets. Extensive experiment results demonstrate that our algorithm outperforms most of the existing aerial image mosaicking methods in visual quality while guaranteeing a high calculation speed.

Mesh-based computation for solving bas-relief reconstruction from editable normal image

As a representative art form between drawing and sculpture, bas-relief is widely used in media for signs, narratives, and decorations. Recent works focus on producing bas-reliefs with fewer interactions and manual efforts. However, they are limited by the large storage size of modeling results, that hinders fast rendering and downstream geometry processing (e.g. bas-relief editing). We propose a mesh-based modeling framework to generate progressive bas-reliefs with different resolutions from editable normal images. Specifically, (1) the normal image is down-sampled to eliminate redundant information but with its visual saliency preserved; (2) to avoid adding any integrability constraints, the editable normal image is changed to mesh representation as the discrete geometry processing setup; (3) surface-from-gradients (SfG) is improved to fit our bas-relief reconstruction with both height and normal constraints. Our method can produce more portable bas-reliefs with smaller sizes for fast visualization, geometry processing, and feeding CNC and 3D printing devices in practice. Experiments show that our method competes favorably with existing methods in visual quality and surface accuracy.

Deep blending for free-viewpoint image-based rendering

Free-viewpoint image-based rendering (IBR) is a standing challenge. IBR methods combine warped versions of input photos to synthesize a novel view. The image quality of this combination is directly affected by geometric inaccuracies of multi-view stereo (MVS) reconstruction and by view- and image-dependent effects that produce artifacts when contributions from different input views are blended. We present a new deep learning approach to blending for IBR, in which we use held-out real image data to learn blending weights to combine input photo contributions. Our Deep Blending method requires us to address several challenges to achieve our goal of interactive free-viewpoint IBR navigation. We first need to provide sufficiently accurate geometry so the Convolutional Neural Network (CNN) can succeed in finding correct blending weights. We do this by combining two different MVS reconstructions with complementary accuracy vs. completeness tradeoffs. To tightly integrate learning in an interactive IBR system, we need to adapt our rendering algorithm to produce a fixed number of input layers that can then be blended by the CNN. We generate training data with a variety of captured scenes, using each input photo as ground truth in a held-out approach. We also design the network architecture and the training loss to provide high quality novel view synthesis, while reducing temporal flickering artifacts. Our results demonstrate free-viewpoint IBR in a wide variety of scenes, clearly surpassing previous methods in visual quality, especially when moving far from the input cameras.

Ink Wash Painting Style Rendering With Physically-based Ink Dispersion Model

This paper presents a real-time rendering method based on the GPU programmable pipeline for rendering the 3D scene in ink wash painting style. The method is divided into main three parts: First, render the ink properties of 3D model by calculating its vertex curvature. Then, cached the ink properties to a paper structure and using an ink dispersion model which is defined by referencing the theory of porous media to simulate the dispersion of ink. Finally, convert the ink properties to the pixel color information and render it to the screen. This method has a better performance than previous methods in visual quality.

Ecological Landscape Planning Considering Landscape Aesthetics (Case Study: Part of Tehran-Qom Freeway)

Visual quality is a significant issue in today’s modern world which needs to be evaluated. This assessment is based on the observer’s perception and sight of space and attempts to introduce the criteria of spatial desirability in the landscape. Analysis of the relationships between visual quality and structural features of the landscape is an effective way of conducting cognitive research. For this purpose, the analysis of visual quality was considered by compiling measurement and evaluation criteria, based on the concept of user preferences, in the part of Tehran-Qom Freeway, near the northeastern side of Imam Khomeini International Airport (Iran). The main objective of this research is to present a strategic plan for preserving and restoring the ecological landscape, creating a sense of place in the region, emphasizing gradual changes in the landscape during the movement and considering its relation to landscape aesthetics. For this purpose, after the questionnaire was prepared, the classifying visual quality method (Q-Sort) was used. Finally, using the results obtained from landscape analysis, identifying the components which led to the lack of visibility of individuals, and by using the geographic information systems (GIS), the strategic design of the area was designed.

Visual Quality Method in Assessing Landscape Characteristics: Case Study of Bozcaada Island

Cengiz, T., 2014. Visual quality method in assessing landscape characteristics: case study of Bozcaada Island.Landscapes are an extremely important resource for tourism and quality of life so there is a need to classify and manage landscapes. In the implementation of visual impact assessment, Bozcaada in the province of Çanakkale was selected as the sample area in the research. An analytical and perceptual method covering physical relationships of user–environment was applied in the research. Initially, the landscape beauty was evaluated according to the survey of observer preferences. Secondly, the landscape elements and attributes were weighted. Finally, regression and correlation analyses were performed to assess the relationships between weighted landscape attributes and survey results. There were 9 panels, each containing 16 photos, and 210 participants ranked the best 4 and worst 4 photos of each panel. The results showed that degree of wilderness, the presence of negative man-made elements, and percentage of land covered by vegetation decreased the visual quality in Bozcaada. Furthermore, it was determined that visual quality increased with the presence of well-preserved man-made elements, scale, harmony, and color contrast. The research revealed which landscape elements and attributes are important in landscape design.