“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

From the Publisher: 
The accessible presentation of this book gives both a general view of the entire computer vision enterprise and also offers sufficient detail to be able to build useful applications. Users learn techniques that have proven to be useful by first-hand experience and a wide range of mathematical methods. A CD-ROM with every copy of the text contains source code for programming practice, color images, and illustrative movies. Comprehensive and up-to-date, this book includes essential topics that either reflect practical significance or are of theoretical importance. Topics are discussed in substantial and increasing depth. Application surveys describe numerous important application areas such as image based rendering and digital libraries. Many important algorithms broken down and illustrated in pseudo code. Appropriate for use by engineers as a comprehensive reference to the computer vision enterprise.

Computer vision : a modern approach = 计算机视觉 : 一种现代的方法

MUCKE aims to mine a large volume of images, to structure them conceptually and to use this conceptual structuring in order to improve large-scale image retrieval. The last decade witnessed important progress concerning low-level image representations. However, there are a number problems which need to be solved in order to unleash the full potential of image mining in applications. The central problem with low-level representations is the mismatch between them and the human interpretation of image content. This problem can be instantiated, for instance, by the incapability of existing descriptors to capture spatial relationships between the concepts represented or by their incapability to convey an explanation of why two images are similar in a content-based image retrieval framework. We start by assessing existing local descriptors for image classification and by proposing to use co-occurrence matrices to better capture spatial relationships in images. The main focus in MUCKE is on cleaning large scale Web image corpora and on proposing image representations which are closer to the human interpretation of images. Consequently, we introduce methods which tackle these two problems and compare results to state of the art methods. Note: some aspects of this deliverable are withheld at this time as they are pending review. Please contact the authors for a preview.

http://ieeexplore.ieee.org/iel2/4524/12820/00592460.pdf

Image Processing

Review of shape representation and description techniques

50 Years of Image Analysis

Moving object segmentation is an essential technique for various video surveillance applications. The result of moving object segmentation often contains shadow regions caused by the color difference of shadow pixels. Hence, moving object segmentation is usually followed by a shadow elimination process to remove the false detection results. The common assumption adopted in previous works is that, under the illumination variation, the value of chromaticity components are preserved while the value of intensity component is changed. Hence, color transforms which separates luminance component and chromaticity component are usually utilized to remove shadow pixels. In this paper, various color spaces (YCbCr, HSI, normalized rgb, Yxy, Lab, c1c2c3) are examined to find the most appropriate color space for shadow elimination. So far, there have been some research efforts to compare the influence of various color spaces for shadow elimination. However, previous efforts are somewhat insufficient to compare the color distortions under illumination change in diverse color spaces, since they used a specific shadow elimination scheme or different thresholds for different color spaces. In this paper, to relieve the limitations of previous works, (1) the amount of gradients in shadow boundaries drawn to uniform colored regions are examined only for chromaticity components to compare the color distortion under illumination change and (2) the accuracy of background subtraction are analyzed via RoC curves to compare different color spaces without the problem of threshold level selection. Through experiments on real video sequences, YCbCr and normalized rgb color spaces showed good results for shadow elimination among various color spaces used for the experiments.

https://www.koreascience.kr:443/article/JAKO200822138618409.pdf

Comparisons of Color Spaces for Shadow Elimination

PROBLEM TO BE SOLVED: To increase an arithmetic processing speed in the case of detecting an object by using only depth informationSOLUTION: The video processing method includes: a step of acquiring the depth information of a video; a step of separating a first region of interest with respect to an area where a target exists from the video; a step of detecting a candidate area with respect to the target from the first region of interest; a step of determining whether or not the detected candidate area corresponds to the target; a step of, when the candidate area corresponds to the target, detecting a target area by using the candidate area; a step of estimating the attitude of the target by using the detected target area; and a step of modeling the target

Video processing method and device

In this paper, we introduce an automatic video management system for personal video recorder With the system, visual data can be summarized and indexed based on the face recognition techniques The developed system also supplies basic summarizing, indexing, and retrieving functionalities for visual data Several conventional algorithms are adapted for cut detection and video summarizing And MPEG-7 visual descriptors are utilized to retrieve similar scene For face detection in real-time, the Haar-like feature method is applied and PCA/LDA method is selected for face recognition The resulting index generates not only a preview of a movie, but also allows non-linear access with thumbnails Moreover, implemented face recognition techniques makes character based video summarization and indexing possible in stored video data

Automatic Visual Data Management System

We present a nearby vehicle detection method in the adjacent lane using in-vehicle front view camera. Nearby vehicles in adjacent lanes show various appearances according to their relative positions to the host vehicle. Therefore, most conventional methods use motion information for detecting nearby vehicles in adjacent lanes. However, these methods can only detect overtaking vehicles which have faster speed than the host vehicle. To solve this problem, we use the feature of regions where nearby vehicle can appear. Consequently, our method cannot only detect nearby overtaking vehicles but also stationary and same speed vehicles in adjacent lanes. In our experiment, we validated our method through various whether, road conditions and real-time implementation.

Nearby Vehicle Detection in the Adjacent Lane using In-vehicle Front View Camera

Shape from focus (SFF), represents a passive technique in recovering object shapes from scenes. In SFF systems, a single imaging device is used to acquire images followed by the application of Focus Measure (FM) at every point to obtain focus quality, to obtain depth by maximizing the focus curves. Although numerous focus measures have been recently proposed, less attention has been paid to the factors affecting them. In regard to FM, one such critical factor impacting SFF system applications is window size selection. In the paper in question, window size criteria for SFF systems have been formulated. The effect of change of imaging device parameters (such as image resolution, focal length of lens etc.) has also been analyzed. Experiments are performed on simulated and real objects to validate the proposed scheme.

Whoi-Yul Kim

Papers

Comparisons of Color Spaces for Shadow Elimination

Video processing method and device

Automatic Visual Data Management System

Nearby Vehicle Detection in the Adjacent Lane using In-vehicle Front View Camera

Calculating Accurate Window Size for Shape-From-Focus