arXiv subjects
Publications and source records attributed to Yongming Zhang.
In this paper, we prove that for any smooth projective curve $C$ of genus $g\geq2$ over an algebraically closed field of positive characteristic, there exists a stable vector bundle over $C$ whose exterior power is not semi-stable.
We begin by formally defining n-Tango curves and n-Raynaud surfaces. Our investigation then focuses on the pathological behaviors exhibited by n-Raynaud surfaces. As a direct corollary of this analysis, we present a concise disproof of Fujita's conjecture for surfaces in positive characteristics.
Creating high-quality anime illustrations presents notable challenges, particularly for beginners, due to the intricate styles and fine details inherent in anime art. We present an interactive drawing guidance system specifically designed for anime illustrations to address this issue. It offers real-time guidance to help users refine their work and streamline the creative process. Our system is built upon the StreamDiffusion pipeline to deliver real-time drawing assistance. We fine-tune Stable Diffusion with LoRA to synthesize anime style RGB images from user-provided hand-drawn sketches and prompts. Leveraging the Informative Drawings model, we transform these RGB images into rough sketches, which are further refined into structured guidance sketches using a custom-designed optimizer. The proposed system offers precise, real-time guidance aligned with the creative intent of the user, significantly enhancing both the efficiency and accuracy of the drawing process. To assess the effectiveness of our approach, we conducted a user study, gathering empirical feedback on both system performance and interface usability.
Deep learning-based sketch-to-clothing image generation provides the initial designs and inspiration in the fashion design processes. However, clothing generation from freehand drawing is challenging due to the sparse and ambiguous information from the drawn sketches. The current generation models may have difficulty generating detailed texture information. In this work, we propose TexControl, a sketch-based fashion generation framework that uses a two-stage pipeline to generate the fashion image corresponding to the sketch input. First, we adopt ControlNet to generate the fashion image from sketch and keep the image outline stable. Then, we use an image-to-image method to optimize the detailed textures of the generated images and obtain the final results. The evaluation results show that TexControl can generate fashion images with high-quality texture as fine-grained image generation.
In this paper, we give a numerical criterion of Reider-type for the $d$-very ampleness of the adjoint line bundles on quasi-elliptic surfaces, and meanwhile we give a new proof of the vanishing theorem on quasi-elliptic surfaces emailed from Langer and show that it is the optimal version.
We present counterexamples to Fujita's conjecture in positive characteristics. Precisely, we show that over any algebraically closed field $k$ of characteristic $p>0$ and for any positive integer $m$, there exists a smooth projective surface $S$ with an ample Cartier divisor $A$ such that the adjoint linear system $|K_S+mA|$ is not free of base point. Our surface $S$ is a certain kind of generalization of Raynaud surfaces.
Let $Y$ be a smooth projective surface defined over an algebraically closed field $k$ with ${\rm Char}\ k\nmid n$, and let $π:X\rightarrow Y$ be a $n$-cyclic covering branched along a smooth divisor $B$. We show that under some conditions $\mathscr{T}_X$ is semi-stable with respect to $π^*\mathcal {H}$ if the tangent bundle $\mathscr{T}_Y$ is semi-stable with respect to some ample line bundle $\mathcal {H}$ on $Y$.
Video smoke detection is a promising fire detection method especially in open or large spaces and outdoor environments. Traditional video smoke detection methods usually consist of candidate region extraction and classification, but lack powerful characterization for smoke. In this paper, we propose a novel video smoke detection method based on deep saliency network. Visual saliency detection aims to highlight the most important object regions in an image. The pixel-level and object-level salient convolutional neural networks are combined to extract the informative smoke saliency map. An end-to-end framework for salient smoke detection and existence prediction of smoke is proposed for application in video smoke detection. The deep feature map is combined with the saliency map to predict the existence of smoke in an image. Initial and augmented dataset are built to measure the performance of frameworks with different design strategies. Qualitative and quantitative analysis at frame-level and pixel-level demonstrate the excellent performance of the ultimate framework.
This paper proposes a method for video smoke detection using synthetic smoke samples. The virtual data can automatically offer precise and rich annotated samples. However, the learning of smoke representations will be hurt by the appearance gap between real and synthetic smoke samples. The existed researches mainly work on the adaptation to samples extracted from original annotated samples. These methods take the object detection and domain adaptation as two independent parts. To train a strong detector with rich synthetic samples, we construct the adaptation to the detection layer of state-of-the-art single-model detectors (SSD and MS-CNN). The training procedure is an end-to-end stage. The classification, location and adaptation are combined in the learning. The performance of the proposed model surpasses the original baseline in our experiments. Meanwhile, our results show that the detectors based on the adversarial adaptation are superior to the detectors based on the discrepancy adaptation. Code will be made publicly available on http://smoke.ustc.edu.cn. Moreover, the domain adaptation for two-stage detector is described in Appendix A.
In this paper, we investigate the problem of simultaneously generations of $r$-jets of $ω_X\otimes L^{\otimes m}$ when $L$ is ample and base point free. It turns out that in this case, the bound of $m$ is optimistic, i.e. $m\geq f(r)=n+r+1$, which is linear in $r$. Our results work over arbitrary characteristics. We also treat the same problem when $X$ is singular where $ω_X$ is replaced by the pushforward of canonical sheaves or the cohomology sheaves of dualizing complexes.
Nonaromatic amino acids are generally believed to be nonemissive, owing to their lack of apparently remarkable conjugation within individual molecules. Here we report the intrinsic visible emission of nonaromatic amino acids and poly(amino acids) in concentrated solutions and solid powders. This unique and widespread luminescent characteristic can be well rationalized by the clustering-triggered emission (CTE) mechanism, namely the clustering of nonconventional chromophores (i.e. amino, carbonyl, and hydroxyl) and subsequent electron cloud overlap with simultaneously conformation rigidification. Such CTE mechanism is further supported by the single crystal structure analysis. Besides prompt fluorescence, room temperature phosphorescence (RTP) are also detected from the solids. Moreover, persistent RTP is observed in the powders of exampled poly(amino acid) of ε-poly-L-lysine (ε-PLL) after ceasing UV irradiation. These results not only illustrate the feasibility of employing the building blocks of nonaromatic amino acids in the exploration of new luminescent biomolecules, but also provide significant implications for the RTP of peptides and proteins at aggregated or crystalline states.
In this paper, a deep domain adaptation based method for video smoke detection is proposed to extract a powerful feature representation of smoke. Due to the smoke image samples limited in scale and diversity for deep CNN training, we systematically produced adequate synthetic smoke images with a wide variation in the smoke shape, background and lighting conditions. Considering that the appearance gap (dataset bias) between synthetic and real smoke images degrades significantly the performance of the trained model on the test set composed fully of real images, we build deep architectures based on domain adaptation to confuse the distributions of features extracted from synthetic and real smoke images. This approach expands the domain-invariant feature space for smoke image samples. With their approximate feature distribution off non-smoke images, the recognition rate of the trained model is improved significantly compared to the model trained directly on mixed dataset of synthetic and real images. Experimentally, several deep architectures with different design choices are applied to the smoke detector. The ultimate framework can get a satisfactory result on the test set. We believe that our approach is a start in the direction of utilizing deep neural networks enhanced with synthetic smoke images for video smoke detection.