Inverse-Like Antagonistic Scene Text Spotting via Reading-Order Estimation and Dynamic Sampling.

Scene text spotting is a challenging task, especially for inverse-like scene text, which has complex layouts, e.g., mirrored, symmetrical, or retro-flexed. In this paper, we propose a unified end-to-end trainable inverse-like antagonistic text spotting framework dubbed IATS, which can effectively spot inverse-like scene texts without sacrificing general ones. Specifically, we propose an innovative reading-order estimation module (REM) that extracts reading-order information from the initial text boundary generated by an initial boundary module (IBM). To optimize and train REM, we propose a joint reading-order estimation loss ( LRE ) consisting of a classification loss, an orthogonality loss, and a distribution loss. With the help of IBM, we can divide the initial text boundary into two symmetric control points and iteratively refine the new text boundary using a lightweight boundary refinement module (BRM) for adapting to various shapes and scales. To alleviate the incompatibility between text detection and recognition, we propose a dynamic sampling module (DSM) with a thin-plate spline that can dynamically sample appropriate features for recognition in the detected text region. Without extra supervision, the DSM can proactively learn to sample appropriate features for text recognition through the gradient returned by the recognition module. Extensive experiments on both challenging scene text and inverse-like scene text datasets demonstrate that our method achieves superior performance both on irregular and inverse-like text spotting.

Fluorinated Carbon Nanohorns as Cathode Materials for Ultra-High Power Li/CFx Batteries.

Fluorinated carbon (CFx) has ultrahigh theoretical energy density among cathode materials for lithium primary batteries. CFx, as an active material in the cathode, plays a decisive role in performance. However, the performance of commercialized fluorinated graphite (FG) does not meet this continuously increasing performance demand. One effective way to increase the overall performance is to manipulate carbon-fluorine (C─F) bonds. In this study, carbon nanohorns are first used as a carbon source and are fluorinated at relatively low temperatures to obtain a new type of CFx with semi-ionic C─F bonds. Carbon nanohorns with a high degree of fluorination achieved a specific capacity comparable to that of commercial FG. Density functional theory (DFT) calculations revealed that curvature structure regulated its C─F bond configuration, thermodynamic parameters, and ion diffusion pathway. The dominant semi-ionic C─F bonds guarantee good conductivity, which improves rate performance. Fluorinated carbon nanohorns delivered a power density of 92.5 kW kg-1 at 50 C and an energy density of 707.6 Wh kg-1 . This result demonstrates the effectiveness of tailored C─F bonds and that the carbon nanohorns shorten the Li+ diffusion path. This excellent performance indicates the importance of designing the carbon source and paves new possibilities for future research.

Time-resolved electron paramagnetic resonance spectrometer based on ultrawide single-sideband phase-sensitive detection.

A time-resolved electron paramagnetic resonance (TREPR) method with 40 ns time resolution and a high sensitivity suitable for the detection of short-lived radicals under thermal equilibrium is developed. The key is the introduction of a new detection technique named ultrawide single sideband phase sensitive detection (U-PSD) to the conventional continuous-wave EPR, which remarkably enhanced the sensitivity for the detection of broadband transient signals compared with the direct detection protocol. By repeatedly triggering a transient kinetic event f(t) (e.g., by laser flash photolysis) under a 100 kHz magnetic field modulation with precise phase control, this technique can build an ultrawide single sideband modulated signal. After single sideband demodulation, the flicker noise-suppressed signal f(t) with wide bandwidth is recovered. A U-PSD TREPR spectrometer prototype has been built, which integrated timing sequence control, laser flash excitation, data acquisition systems, and the U-PSD algorithm with a conventional continuous-wave EPR. It exhibited excellent performance in monitoring a model transient radical system, laser flash photolysis of benzophenone in isopropanol. Both the intense chemically induced dynamic electron polarization signals and the much weaker thermal equilibrium EPR signals of the generated acetone ketyl radical and benzophenone ketyl radical were clearly observed within a wide timescale ranging from sub-microsecond to milliseconds. This prototype validated the feasibility of the U-PSD technique and demonstrated its superior performance in studying complex photochemical systems containing various transient radicals, which complements the established TREPR techniques and provides a powerful tool for deep mechanistic understandings, such as in photoredox catalysis and artificial photosynthesis.

Arbitrary Shape Text Detection via Segmentation With Probability Maps.

Arbitrary shape text detection is a challenging task due to the significantly varied sizes and aspect ratios, arbitrary orientations or shapes, inaccurate annotations, etc. Due to the scalability of pixel-level prediction, segmentation-based methods can adapt to various shape texts and hence attracted considerable attention recently. However, accurate pixel-level annotations of texts are formidable, and the existing datasets for scene text detection only provide coarse-grained boundary annotations. Consequently, numerous misclassified text pixels or background pixels inside annotations always exist, degrading the performance of segmentation-based text detection methods. Generally speaking, whether a pixel belongs to text or not is highly related to the distance with the adjacent annotation boundary. With this observation, in this paper, we propose an innovative and robust segmentation-based detection method via probability maps for accurately detecting text instances. To be concrete, we adopt a Sigmoid Alpha Function (SAF) to transfer the distances between boundaries and their inside pixels to a probability map. However, one probability map can not cover complex probability distributions well because of the uncertainty of coarse-grained text boundary annotations. Therefore, we adopt a group of probability maps computed by a series of Sigmoid Alpha Functions to describe the possible probability distributions. In addition, we propose an iterative model to learn to predict and assimilate probability maps for providing enough information to reconstruct text instances. Finally, simple region growth algorithms are adopted to aggregate probability maps to complete text instances. Experimental results demonstrate that our method achieves state-of-the-art performance in terms of detection accuracy on several benchmarks. Notably, our method with Watershed Algorithm as post-processing achieves the best F-measure on Total-Text (88.79%), CTW1500 (85.75%), and MSRA-TD500 (88.93%). Besides, our method achieves promising performance on multi-oriented datasets (ICDAR2015) and multilingual datasets (ICDAR2017-MLT). Code is available at: https://github.com/GXYM/TextPMs.

Kernel Proposal Network for Arbitrary Shape Text Detection.

Segmentation-based methods have achieved great success for arbitrary shape text detection. However, separating neighboring text instances is still one of the most challenging problems due to the complexity of texts in scene images. In this article, we propose an innovative kernel proposal network (dubbed KPN) for arbitrary shape text detection. The proposed KPN can separate neighboring text instances by classifying different texts into instance-independent feature maps, meanwhile avoiding the complex aggregation process existing in segmentation-based arbitrary shape text detection methods. To be concrete, our KPN will predict a Gaussian center map for each text image, which will be used to extract a series of candidate kernel proposals (i.e., dynamic convolution kernel) from the embedding feature maps according to their corresponding keypoint positions. To enforce the independence between kernel proposals, we propose a novel orthogonal learning loss (OLL) via orthogonal constraints. Specifically, our kernel proposals contain important self-information learned by network and location information by position embedding. Finally, kernel proposals will individually convolve all embedding feature maps for generating individual embedded maps of text instances. In this way, our KPN can effectively separate neighboring text instances and improve the robustness against unclear boundaries. To the best of our knowledge, our work is the first to introduce the dynamic convolution kernel strategy to efficiently and effectively tackle the adhesion problem of neighboring text instances in text detection. Experimental results on challenging datasets verify the impressive performance and efficiency of our method. The code and model are available at https://github.com/GXYM/KPN.

Laparoscopic Duodenal Switch Versus Roux-en-Y Gastric Bypass at a High-Volume Community Hospital: a Retrospective Cohort Study from a Rural Setting.

BACKGROUND: The classic duodenal switch (DS) represents a minority of bariatric procedures due to its high complexity and potential for complications. METHODS: A retrospective chart review was conducted on 100 laparoscopic DS cases from 2014 to 2018 at an accredited program in a rural community hospital and compared to 100 laparoscopic Roux-en-Y gastric bypasses (RYGB). Primary outcomes were 30-day morbidity and mortality. Secondary outcomes included anastomotic leak and remission of type 2 diabetes. RESULTS: There were more demographic risk factors for DS. The 30-day morbidity was higher for DS compared to RYGB (31% vs 13%, respectively; p = 0.0037). There was one mortality for DS and none for RYGB. There were statistically significant longer intraoperative times, greater EBL, and greater decrease in BMI for DS. The DS had a lower incidence of anastomotic ulcers (4% vs 13%, respectively; p = 0.0289), with a higher incidence of subsequent surgery beyond 30 days (21% vs 8%, respectively; p = 0.0160). There were 3 anastomotic leaks for DS and none for RYGB, although not statistically significant (p = 0.2463). The DS was more likely to eradicate hypertension, but the RYGB was more likely to eradicate GERD. There were no statistically significant differences for type 2 diabetes remission (92.1% vs 89.5%, respectively; p = 0.7239). CONCLUSION: Laparoscopic DS offers greater weight loss and hypertension remission, with lower incidence of anastomotic ulcers, but at the expense of greater morbidity and need for subsequent surgery, with no significant differences in type 2 diabetes remission when compared to RYGB in a rural community hospital program.

Improved generation method utilizing a modified Fourier spectrum for Airy beams with the phase-only filter technique.

We present an improved method to generate Airy beams utilizing a liquid crystal on silicon (LCoS) device. In this method, the phase and amplitude information of a modified Fourier spectrum of an Airy beam together with a Fresnel holographic lens is encoded onto the LCoS using the phase-only filter technique; thus, a desired Airy beam is formed in the focal plane of the Fresnel holographic lens. In this paper, the principle of the proposed method is described in detail, and both the excellent numerical simulations and experimental results for verifying this method are demonstrated. It is shown that the new generation method is accurate and simple; in particular, the setup is more compact compared to the conventional Fourier transform method, which comprises only the input polarized laser and a LCoS device. This effective method will further promote investigations into the properties and applications of Airy beams.