Contemporary display enables video content rendering with high dynamic range (HDR) and wide color gamut (WCG). However, the majority of existing content remains in standard dynamic range (SDR) format. Therefore, the conversion of SDR content to HDRTV standards holds significant value. This paper delineates and analyzes the SDRTV-to-HDRTV conversion by modeling the formation of SDRTV/HDRTV content. The findings reveal that a naive end-to-end supervised training pipeline suffers from severe gamut transition errors. To address this, we propose a new three-step solution called HDRTVNet++, which includes adaptive global color mapping, local enhancement, and highlight refinement. The adaptive global color mapping step utilizes global statistics for image-adaptive color adjustments, followed by a local enhancement network for detail improvement. These two components are integrated as a generator, with GAN-based joint training ensuring highlight consistency. Our method, tailored for ultra-high-definition TV content, offers both effectiveness and computational efficiency in processing 4K resolution images. We also construct HDRTV1K, a dataset comprising HDR videos adhering to the HDR10 standard, featuring 1235 training and 117 testing images at 4K resolution. Furthermore, we employ five metrics to assess SDRTV-to-HDRTV performance. Our results demonstrate state-of-the-art performance both quantitatively and visually. The codes and models are available athttps://github.com/xiaom233/HDRTVNet-plus.

Photo retouching aims at improving the aesthetic visual quality of images that suffer from photographic defects, especially for poor contrast, over/under exposure, and inharmonious saturation. In practice, photo retouching can be accomplished by a series of image processing operations. As most commonly-used retouching operations are pixel-independent, i.e., the manipulation on one pixel is uncorrelated with its neighboring pixels, we can take advantage of this property and design a specialized algorithm for efficient global photo retouching. We analyze these global operations and find that they can be mathematically formulated by a Multi-Layer Perceptron (MLP). Based on this observation, we propose an extremely lightweight framework – Conditional Sequential Retouching Network (CSRNet). Benefiting from the utilization of$1\times 1$convolution, CSRNet only contains less than 37 K trainable parameters, which are orders of magnitude smaller than existing learning-based methods. Experiments show that our method achieves state-of-the-art performance on the benchmark MIT-Adobe FiveK dataset quantitively and qualitatively. In addition to achieve global photo retouching, the proposed framework can be easily extended to learn local enhancement effects. The extended model, namely CSRNet-L, also achieves competitive results in various local enhancement tasks.

The methods of Bayesian predictive inference and variation index are vital in the constant false alarm rate (CFAR) detection, which adaptively predicts and evaluates the background level to achieve target detection. However, in complex scenarios, the interference effect can lead to imprecise prediction and complex calculations of the background level, which results in severely degraded CFAR performance. To solve the above situation, a robust variation index CFAR detector based on Bayesian interference control (BVI-CFAR) is proposed. The clutter range profile (CRP) is equally divided and the decision-making selection is optimized to dynamically evaluate the background clutter level in this algorithm; meanwhile, the clutter level will feedback control the former. Then, based on the Bayesian interference control theory, the Bayesian classification interference control idea is proposed to simultaneously predict inference multiple targets in the background. Therefore, the anti-interference ability of the detector is improved while reducing the computational complexity. The false alarm rate and decision expression are given, and the detection is extended when the segmentation and interference are arbitrary. Both the simulation results and the field test results verify the favorable performance of the algorithm in complex scenarios.

Nowadays modern displays are capable to render video content with high dynamic range (HDR) and wide color gamut (WCG). However, most available resources are still in standard dynamic range (SDR). Therefore, there is an urgent demand to transform existing SDR-TV contents into their HDR-TV versions. In this paper, we conduct an analysis of SDRTV-to-HDRTV task by modeling the formation of SDRTV/HDRTV content. Base on the analysis, we propose a three-step solution pipeline including adaptive global color mapping, local enhancement and highlight generation. Moreover, the above analysis inspires us to present a lightweight network that utilizes global statistics as guidance to conduct image-adaptive color mapping. In addition, we construct a dataset using HDR videos in HDR10 standard, named HDRTV1K, and select five metrics to evaluate the results of SDRTV-to-HDRTV algorithms. Furthermore, our final results achieve state-of-the-art performance in quantitative comparisons and visual quality. The code and dataset are available at https://github.com/chxy95/HDRTVNet.

Although deep neural networks (DNNs) have achieved excellent generalization performance on well-labelled datasets, the real-world training sets usually suffer from label noise. Owing to the powerful representational ability, DNNs may be vulnerable to the noisy data thus resulting in a sharp deterioration of test performance. To tackle this problem, we propose a novel cross-training framework to leverage two synergic deep architectures, where each member plays the roles of both teacher of generating curriculums and student of updating model parameters via a novel loss function. It is convenient to train our proposed model directly without any prior information such as noise rate. The state-of-the-art experimental results on several benchmark datasets under different types of label noise have demonstrated the effectiveness and robustness of our method.

Semisupervised incremental learning is the task of classifying data streams with partially labeled data when annotation information is difficult to obtain. Besides the sequential learning manner and lack of label information, multiple novel classes and concept drift may emerge from incremental learning. Most previous studies have only considered these problems in part. To tackle challenges involved in semisupervised incremental learning, an adaptive matrix sketching and clustering method is proposed in this letter, which cohesively and adaptively classifies known classes, identifies multiple novel classes, and updates the learning model. Experiments were conducted to evaluate this method on three benchmark datasets, containing various data types, including network attack analysis, the geospatial information of forests, and images of handwritten numbers. Results validated the effectiveness of our proposed method and its superiority over many previous studies.

Along with the fast-growth in both scale and frequency of academic conferences and advances in communication and information technologies, digital revolutions are happening to conferencing affairs. Paper review assignment is a hard and time-consuming job due to many constraints, even emotional factors affect the review assignment. In this paper, the human-knowledge network, aggregated based on the past conference publications, social networks of PC members' or practical reviewers' and their expertise, is adopted into paper review assignment to tackle those constraints so as to generate an integer programming model to produce optimal alternatives for practical assignment..

A trial approach is presented for the realization of an optical character reader for printed Chinese characters (kanji). This approach is different from conventional ones as far as kanji features and recognition are concerned. Use is made of structural features based on only the horizontal and vertical strokes in a kanji, taking advantage of optical/digital hybrid processing. The recognition experimental results are promising: 99.6%, 0.3%, and 0.1% for the correct, rejected, and error rates, respectively.>