Efficient image retargeting with Bezier curves

With the increasing requirement of efficient image retargeting, many algorithms have been proposed for adapting images contents to various display settings. However, most of these algorithms work in spatial domain of raw images. Since images are mostly stored in DCT-based compressed format such as JPEG, it will be very attractive to realize the image retargeting in compressed domain. In this paper, we propose a new low complexity DCT-based image retargeting method, which is completely performed in compressed domain. This proposed algorithm is based on the construction of block level importance map and calculation of block level forward energy. Experimental results prove that our proposed algorithm is able to significantly reduce image decoding complexity as well as image retargeting complexity, while providing the satisfying image retargeting quality compared with other methods.

Effective and Efficient retargeting are critical to improve user browsing experiences in mobile devices. One important issue in previous works lies in their semantic gap in modeling user focuses and intensions from low-level features, which results to data noise in their importance map constructions. Towards noise-tolerance learning for effective retargeting, we propose a generalized content aware framework from a supervised learning viewpoint. Our main idea is to revisit the retargeting process as working out an optimal mapping function to approximate the output (desirable pixel-wise or region-wise changes) from the training data. Therefore, we adopt a prediction error decomposition strategy to measure the effectiveness of the previous retargeting methods. In addition, taking into account the data noise in importance maps, we also propose a grid-based retargeting model, which is robust and effective to data noise in real time retargeting function learning. Finally, using different mapping functions, our framework is generalized for explaining previous works, such as seam carving [9,13] and mesh based methods [3,18]. Extensive experimental comparison to state-of-the-art works have shown promising results of the proposed framework.

A fast hybrid retargeting scheme with seam context and content aware strip partition

Image registration and retargeting has been playing a significant role in creating sorts of visual effects in animation production pipeline, attracting plenty of researchers working on. However, previous methods are slow due to physically or geometrically based complicated computations and often rely on user assistants or external devices to capture the motion data, reducing their practicability. In this paper, we propose a novel approach to capture a path motion contained in a source animation in a fully automatic manner and reuse it to new characters with different appearances. The key idea is to employ visual difference based random search procedure to automatically track the motion and combine the moving least square based deformation technique to endue the new object with the captured motion. Our experiments demonstrate that our method achieves good results with high quality and performance.

This paper proposes a novel image-retargeting algorithm that can retarget images to a large family of non-rectangular shapes. Specifically, we study image retargeting from a broader perspective that includes the content as well as the shape of an image, and the proposed content and shape-aware image-retargeting (CASAIR) algorithm is driven by the dual objectives of image content preservation and image domain transformation, with the latter defined by an application-specific target shape. The algorithm is based on the idea of seam segment carving that successively removes low-cost seam segments from the image to simultaneously achieve the two objectives, with the selection of seam segments determined by a cost function incorporating inputs from image content and target shape. To provide a complete characterization of shapes that can be obtained using CASAIR, we introduce the notion of bhv-convex shapes, and we show that bhv-convex shapes are precisely the family of shapes that can be retargeted to by CASAIR. The proposed algorithm is simple in both its design and implementation, and in practice, it offers an efficient and effective retargeting platform that provides its users with considerable flexibility in choosing target shapes. To demonstrate the potential of CASAIR for broadening the application scope of image retargeting, this paper also proposes a smart camera-projector system that incorporates CASAIR. In the context of ubiquitous display, CASAIR equips the camera-projector system with the capability of retargeting images online in order to maximize the quality and fidelity of the displayed images whenever the situation demands.

Several imaging and graphics applications usually require the resizing/retargeting of the spatial resolution of a given image. In real time applications, the retargeting process is expected to be computationally fast and free of any visually disturbing structural distortions. Accordingly, several content-aware retargeting methods, which resize images while preserving important image structures, have been introduced. However, it has also been shown that finding a single method that can be effectively applicable in all types of situation is very challenging. Specifically, many of the existing methods are not capable of avoiding visible structural distortions during high level of image squeezing or stretching processes. The distortions become even more evident with the presence of numerous high frequency contents. Accordingly, we have proposed a new and efficient image retargeting method which is based on Radial Basis Function (RBF) interpolation. We have introduced a novel pre-processing step and modifications to the generic form of the RBF interpolation technique for a better structural as well as size preservation of important image regions. The method is finally evaluated with respect to prior retargeting methods and leads to improved performances, both in terms of quality and complexity measures.

. A typical target recognition system usually involves three stages: image enhancement, image segmentation, and object identification. We propose an efficient image segmentation approach to segment man-made targets from unmanned aerial vehicle (UAV) imagery using curvature information derived from image histogram smoothed by Bezier splines. The experimental results show that by enhancing the histogram instead of the original image, similar segmentation results can be obtained in a more efficient way. Algorithm complexity is analyzed. Segmentation results from images enhanced by variants of diffusion methods are also presented for comparison. Real-time automatic target recognition (aTr) is the main objective of the proposed approach.

Small scale quad-rotor unmanned aerial vehicle (UAV) has attracted much attention in recent years and has been widely adopted in many civil applications, e.g. inspection of critical infrastructure spanning over a large geographical area. In a typical UAV based inspection system for large-scale photovoltaic farm, it is required to control the mounted gimbal camera taking pictures of all the PV modules and eventually accurately identify the ones with defects through efficient image processing techniques. Given a series of waypoints without an effective path generation algorithm, the image acquisition can be not satisfactory and even miss some areas that should be inspected. In this paper, a Bezier curve and particle swarm optimization (PSO) joint approach is presented to address the UAV path planning challenge. The path generation process fully takes flight attitude, gimbal limitation and path length into consideration with the aim to improve the efficiency and the reliability of the inspection system. The performance of the proposed solution is assessed through a set of simulation experiments and the numerical result demonstrates the effectiveness of the proposed optimal path planning solution.

Most of the real world scenes have a very high dynamic range (HDR). The mobile phone cameras and the digital cameras available in markets are limited in their capability in both the range and spatial resolution. Same argument can be posed about the limited dynamic range display devices which also differ in the spatial resolution and aspect ratios. In this paper, we address the problem of displaying the high contrast low dynamic range (LDR) image of a HDR scene in a display device which has different spatial resolution compared to that of the capturing digital camera. The optimal solution proposed in this work can be employed with any camera which has the ability to shoot multiple differently exposed images of a scene. Further, the proposed solutions provide the flexibility in the depiction of entire contrast of the HDR scene as a LDR image with an user specified spatial resolution. This task is achieved through an optimized content aware retargeting framework which preserves salient features along with the algorithm to combine multi-exposure images. We show the proposed approach performs exceedingly well in the generation of high contrast LDR image of varying spatial resolution compared to an alternate approach.

Due to the rapid growth of digital gadgets with various screen sizes, resolutions and hardware processing capabilities, robust video retargeting is of increasing relevance. An efficient retargeting algorithm should not only retain semantic content, but also maintain spatiotemporal resolution of video data. In this paper, the effective seam carving technique for content-aware video retargeting is discussed. Retargeting video is of immense importance as it is frequently played on several gadgets such as television, mobile, tablet, and notebook. The proposed method considers each video frame as an independent image entity and tries to resize it. Our main contribution is a formulation of seam carving using graph cut method. Convention cut techniques fail to defend a meaningful seam. Single monotonic well connected by pixel to pixel is most desirable property in seam carving process. The traditional seam carving method is designed to work based on the minimum energy concept, while ignoring the energy that has been introduced by the operator. To address this issue, we propose a new design criterion in which least amount of energy is introduced in retargeted video.

Image retargeting is one of the most popular multimedia content and manipulation technique in the Computer Science world. In this paper, we find the solution for shadow image efficient retargeting technique. The proposed technology is called i-CRIST (improved Content Retargeting Image Seam Carving Technique). The method is a combination of two resizing Methods Region based resizing and background seam carving. The first method identifies the shadow objects in the input image and performs Region based retargeting. Second method seam carving works all non-shadow (background) regions. The improved CRIST performs much efficiently on shadow images then the state of art retargeting technique. We evaluate i-CRIST method on shadow image dataset in [24]. The experiment result of proposed system outperforms rest of all retargeting methodologies. The Result accuracy is calculated based on Resolution and Mean Opinion Score [4] (MOS).

A fully-labeled image dataset provides an exclusive resource for reproducible analysis, investigation inquiries and data analyses in different research computational fields like machine learning, computer vision and deep learning machine intelligence. This research paper present a large scale fully-labeled natural image dataset for Multi-Operator content aware image retargeting techniques. The image dataset is feely available for image processing research field. The current research natural image dataset entitled CRIST900, it include 900 natural images and uses for content aware image retargeting. The proposed CRIST is an image retargeting Multi-Operator method called Content Retargeting Image reSizing Technique. The proposed image resizing method has three phases, image object boundary identification, the image objects feature importance visual saliency map generation and Multi-Operator techniques for efficient retargeting of image objects. The Multi-Operator retargeted image quality assessment was done by different subjective and objective image quality assessment matrices. The experimental result shows that the proposed approach can attain better result than the traditional methods and content aware state-of-the-art image retargeting techniques. The research dataset is publicly available at https://sites.google.com/view/abhayadevmalayil/home.

Nowadays there are many types of display devices with different resolutions and device adaptation is a great challenge in networking. This requires a powerful method for fast resize of images with acceptable quality. Amongst all available methods, such as cropping or scaling, Seam Carving (SC) is a simple and efficient content aware image resizing technique. This technique is inherently a sequential process, which translates into long execution time. In this paper we improve SC by proposing a trellis-based method that finds and removes multiple non-conflicting seams. Better preservation of visually important contents of an image is also achieved by using a new saliency measure called noticeability map. Experiments showed that this approach is up to 13.8 times faster than the original SC, while the produced images are of the same or better visual quality comparing to the original SC or similar methods.

Several different methods have been proposed for image/video retargeting while retaining the content. However, they sometimes produce some artifacts, such as ridge or structure twist. In this paper, we present a structure-preserving image resizing technique for the image retargeting applications. Based on the warping-based retargeting technique proposed by Wolf et al.[13], we propose an efficient and adaptive image resizing algorithm that preserves the content and image structure as best as possible. We first downsample the size of the original image by using bilinear interpolation. In order to preserve the content, we introduce the structure constraints derived from the line detection into the large linear system. Then, the mapping matrices are enlarged to the original size by joint-bilateral upsampling and the resized image can be produced to preserve the content and structure as best as possible. Most of the computation is on the low-resolution layer and therefore it can be very efficient. From our experiments, the proposed method can provide resized images with higher image quality and faster speed than that in [13].

In this paper, an efficient and adjustable visual image encryption scheme is proposed by combining a 6D hyperchaotic system, compressive sensing, and Bezier curve embedding. First, the plain image is sparse by discrete wavelet transform (DWT). Then, the sparse image is encrypted and compressed through game-of-life (GOL) hybrid scrambling and compressive sensing into a cipher image. Next, Bezier curve embedding is utilized to embed the cipher image into the carrier image in wavelet domain. After these operations, the final visually meaningful steganographic image is generated. Additionally, the frequency-domain information of the plain image is used to generate the initial values of the 6D hyperchaotic system in scrambling process, which makes the proposed encryption scheme able to effectively resist the chosen-plaintext attacks (CPA) and the known-plaintext attacks (KPA). Moreover, our scheme exhibits excellent adjustable performance compared with existing related schemes. Ultimately, simulation results and comprehensive performance analyses demonstrate that the scheme proposed in this paper has high decryption quality, visual security, robustness, and operating efficiency.