Numerous single image deraining algorithms have been recently proposed. However, these algorithms are mainly evaluated using certain type of synthetic images, assuming a specific rain model, plus a few real images. It is thus unclear how these algorithms would perform on rainy images acquired "in the wild" and how we could gauge the progress in the field. This paper aims to bridge this gap. We present a comprehensive study and evaluation of existing single image deraining algorithms, using a new largescale benchmark consisting of both synthetic and realworld rainy images of various rain types. This dataset highlights diverse rain models (rain streak, rain drop, rain and mist), as well as a rich variety of evaluation criteria (fulland no-reference objective, subjective, and task-specific) Our evaluation and analysis indicate the performance gap between synthetic rainy images and real-world images and allow us to better identify the strengths and limitations of each method as well as future research directions.

BackgroundCoffee is one of the world's most important crops; it is consumed worldwide and plays a significant role in the economy of producing countries. Coffea arabica and C. canephora are responsible for 70 and 30% of commercial production, respectively. C. arabica is an allotetraploid from a recent hybridization of the diploid species, C. canephora and C. eugenioides. C. arabica has lower genetic diversity and results in a higher quality beverage than C. canephora. Research initiatives have been launched to produce genomic and transcriptomic data about Coffea spp. as a strategy to improve breeding efficiency.ResultsAssembling the expressed sequence tags (ESTs) of C. arabica and C. canephora produced by the Brazilian Coffee Genome Project and the Nestlé-Cornell Consortium revealed 32,007 clusters of C. arabica and 16,665 clusters of C. canephora. We detected different GC3 profiles between these species that are related to their genome structure and mating system. BLAST analysis revealed similarities between coffee and grape (Vitis vinifera) genes. Using KA/KS analysis, we identified coffee genes under purifying and positive selection. Protein domain and gene ontology analyses suggested differences between Coffea spp. data, mainly in relation to complex sugar synthases and nucleotide binding proteins. OrthoMCL was used to identify specific and prevalent coffee protein families when compared to five other plant species. Among the interesting families annotated are new cystatins, glycine-rich proteins and RALF-like peptides. Hierarchical clustering was used to independently group C. arabica and C. canephora expression clusters according to expression data extracted from EST libraries, resulting in the identification of differentially expressed genes. Based on these results, we emphasize gene annotation and discuss plant defenses, abiotic stress and cup quality-related functional categories.ConclusionWe present the first comprehensive genome-wide transcript profile study of C. arabica and C. canephora, which can be freely assessed by the scientific community at http://www.lge.ibi.unicamp.br/coffea. Our data reveal the presence of species-specific/prevalent genes in coffee that may help to explain particular characteristics of these two crops. The identification of differentially expressed transcripts offers a starting point for the correlation between gene expression profiles and Coffea spp. developmental traits, providing valuable insights for coffee breeding and biotechnology, especially concerning sugar metabolism and stress tolerance.

No abstract

The recent computer graphics developments have upraised the quality of the generated digital content, astonishing the most skeptical viewer. Games and movies have taken advantage of this fact but, at the same time, these advances have brought serious negative impacts like the ones yielded by fake images produced with malicious intents. Digital artists can compose artificial images capable of deceiving the great majority of people, turning this into a very dangerous weapon in a timespan currently know as "Fake News/Post-Truth" Era. In this work, we propose a new approach for dealing with the problem of detecting computer generated images, through the application of deep convolutional networks and transfer learning techniques. We start from Residual Networks and develop different models adapted to the binary problem of identifying if an image was or not computer generated. Differently from the current state-of-the-art approaches, we don't rely on hand-crafted features, but provide to the model the raw pixel information, achieving the same 0.97 of state-of-the-art methods with two main advantages: our methods show more stable results (depicted by lower variance) and eliminate the laborious and manual step of specialized features extraction and selection.

The capability of image deraining is a highly desirable component of intelligent decision-making in autonomous driving and outdoor surveillance systems. Image deraining aims to restore the clean scene from the degraded image captured in a rainy day. Although numerous single image deraining algorithms have been recently proposed, these algorithms are mainly evaluated using certain type of synthetic images, assuming a specific rain model, plus a few real images. It remains unclear how these algorithms would perform on rainy images acquired "in the wild" and how we could gauge the progress in the field. This paper aims to bridge this gap. We present a comprehensive study and evaluation of existing single image deraining algorithms, using a new large-scale benchmark consisting of both synthetic and real-world rainy images of various rain types. This dataset highlights diverse rain models (rain streak, rain drop, rain and mist), as well as a rich variety of evaluation criteria (full-and no-reference objective, subjective, and task-specific). We further provide a comprehensive suite of criteria for deraining algorithm evaluation, including full-and no-reference metrics, subjective evaluation, and the novel task-driven evaluation. The proposed benchmark is accompanied with extensive experimental results that facilitate the assessment of the state-of-the-arts on a quantitative basis. Our evaluation and analysis indicate the gap between the achievable performance on synthetic rainy images and the practical demand on real-world images. We show that, despite many advances, image deraining is still a largely open problem. The paper is concluded by summarizing our general observations, identifying open research challenges and pointing out future directions. Our code and dataset is publicly available at http://uee.me/ddQsw. Keywords Image deraining • Image quality assessment • Deep convolution network • Benchmark analysis Communicated by Torsten Sattler.