Search

Your search keyword '"Sun, Xuecheng"' showing total 598 results
Start Over Author "Sun, Xuecheng"
598 results on '"Sun, Xuecheng"'

7. Prompt-based test-time real image dehazing: a novel pipeline

Author

Chen, Zixuan, He, Zewei, Lu, Ziqian, Sun, Xuecheng, and Lu, Zhe-Ming

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Existing methods attempt to improve models' generalization ability on real-world hazy images by exploring well-designed training schemes (\eg, CycleGAN, prior loss). However, most of them need very complicated training procedures to achieve satisfactory results. For the first time, we present a novel pipeline called Prompt-based Test-Time Dehazing (PTTD) to help generate visually pleasing results of real-captured hazy images during the inference phase. We experimentally observe that given a dehazing model trained on synthetic data, fine-tuning the statistics (\ie, mean and standard deviation) of encoding features is able to narrow the domain gap, boosting the performance of real image dehazing. Accordingly, we first apply a prompt generation module (PGM) to generate a visual prompt, which is the reference of appropriate statistical perturbations for mean and standard deviation. Then, we employ a feature adaptation module (FAM) into the existing dehazing models for adjusting the original statistics with the guidance of the generated prompt. PTTD is model-agnostic and can be equipped with various state-of-the-art dehazing models trained on synthetic hazy-clean pairs to tackle the real image dehazing task. Extensive experimental results demonstrate that our PTTD is effective, achieving superior performance against state-of-the-art dehazing methods in real-world scenarios. The code is available at \url{https://github.com/cecret3350/PTTD-Dehazing}., Comment: Accepted by ECCV 2024

Published
2023

8. Accurate and lightweight dehazing via multi-receptive-field non-local network and novel contrastive regularization

Author

He, Zewei, Chen, Zixuan, Lu, Ziqian, Sun, Xuecheng, and Lu, Zhe-Ming

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Recently, deep learning-based methods have dominated image dehazing domain. Although very competitive dehazing performance has been achieved with sophisticated models, effective solutions for extracting useful features are still under-explored. In addition, non-local network, which has made a breakthrough in many vision tasks, has not been appropriately applied to image dehazing. Thus, a multi-receptive-field non-local network (MRFNLN) consisting of the multi-stream feature attention block (MSFAB) and cross non-local block (CNLB) is presented in this paper. We start with extracting richer features for dehazing. Specifically, we design a multi-stream feature extraction (MSFE) sub-block, which contains three parallel convolutions with different receptive fields (i.e., $1\times 1$, $3\times 3$, $5\times 5$) for extracting multi-scale features. Following MSFE, we employ an attention sub-block to make the model adaptively focus on important channels/regions. The MSFE and attention sub-blocks constitute our MSFAB. Then, we design a cross non-local block (CNLB), which can capture long-range dependencies beyond the query. Instead of the same input source of query branch, the key and value branches are enhanced by fusing more preceding features. CNLB is computation-friendly by leveraging a spatial pyramid down-sampling (SPDS) strategy to reduce the computation and memory consumption without sacrificing the performance. Last but not least, a novel detail-focused contrastive regularization (DFCR) is presented by emphasizing the low-level details and ignoring the high-level semantic information in the representation space. Comprehensive experimental results demonstrate that the proposed MRFNLN model outperforms recent state-of-the-art dehazing methods with less than 1.5 Million parameters., Comment: submitted to IEEE TCYB for possible publication

Published
2023

14. Highly stable and antifungal properties on the oilseed rape of Cu3(MoO4)2(OH)2 nanoflakes prepared by simple aqueous precipitation

Author

Zhao Xu, Xu Lisha, Liu Yi, Mei Yunjun, Chen Luocheng, Zheng Anqi, Yin Kuibo, Xiao Xiaolu, Li Shaozhen, Sun Xuecheng, and Zhang Yifu

Subjects
Medicine, Science
Abstract

Abstract In the last few decades, nanoparticles have been a prominent topic in various fields, particularly in agriculture, due to their unique physicochemical properties. Herein, molybdenum copper lindgrenite Cu3(MoO4)2(OH)2 (CM) nanoflakes (NFs) are synthesized by a one-step reaction involving α-MoO3 and CuCO3⋅Cu(OH)2⋅xH2O solution at low temperature for large scale industrial production and developed as an effective antifungal agent for the oilseed rape. This synthetic method demonstrates great potential for industrial applications. Infrared spectroscopy and X-ray diffraction (XRD) results reveal that CM samples exhibit a pure monoclinic structure. TG and DSC results show the thermal stable properties. It can undergo a phase transition form copper molybdate (Cu3Mo2O9) at about 300 °C. Then Cu3Mo2O9 nanoparticles decompose into at CuO and MoO3 at 791 °C. The morphology of CM powder is mainly composed of uniformly distributed parallelogram-shaped nanoflakes with an average thickness of about 30 nm. Moreover, the binding energy of CM NFs is measured to be 2.8 eV. To assess the antifungal properties of these materials, both laboratory and outdoor experiments are conducted. In the pour plate test, the minimum inhibitory concentration (MIC) of CM NFs against Sclerotinia sclerotiorum (S. sclerotiorum) is determined to be 100 ppm, and the zone of inhibiting S. sclerotiorum is 14 mm. When the concentration is above 100 nm, the change rate of the hyphae circle slows down a little and begins to decrease until to 200 ppm. According to the aforementioned findings, the antifungal effects of a nano CM NFs solution are assessed at different concentrations (0 ppm (clear water), 40 ppm, and 80 ppm) on the growth of oilseed rape in an outdoor setting. The results indicate that the application of CM NFs led to significant inhibition of S. sclerotiorum. Specifically, when the nano CM solution was sprayed once at the initial flowering stage at a concentration of 80 ppm, S. sclerotiorum growth was inhibited by approximately 34%. Similarly, when the solution was sprayed once at the initial flowering stage and once at the rape pod stage, using a concentration of 40 ppm, a similar level of inhibition was achieved. These outcomes show that CM NFs possess the ability to bind with more metal ions due to their larger specific surface area. Additionally, their semiconductor physical properties enable the generation of reactive oxygen species (ROS). Therefore, CM NFs hold great potential for widespread application in antifungal products.

Published
2024
Full Text
View/download PDF

38. DuSiamIE: A Lightweight Multidimensional Infrared-Enhanced RGBT Tracking Algorithm for Edge Device Deployment.

Author

Li, Jiao, Wu, Haochen, Gu, Yuzhou, Lu, Junyu, and Sun, Xuecheng

Abstract

Advancements in deep learning and infrared sensors have facilitated the integration of RGB-thermal (RGBT) tracking technology in computer vision. However, contemporary RGBT tracking methods handle complex image data, resulting in inference procedures with a large number of floating-point operations and parameters, which limits their performance on general-purpose processors. We present a lightweight Siamese dual-stream infrared-enhanced RGBT tracking algorithm, called DuSiamIE.It is implemented on the low-power NVIDIA Jetson Nano to assess its practicality for edge-device applications in resource-limited settings. Our algorithm replaces the conventional backbone network with a modified MobileNetV3 and incorporates light-aware and infrared feature enhancement modules to extract and integrate multimodal information. Finally, NVIDIA TensorRT is used to improve the inference speed of the algorithm on edge devices. We validated our algorithm on two public RGBT tracking datasets. On the GTOT dataset, DuSiamIE achieved a precision (PR) of 83.4% and a success rate (SR) of 66.8%, with a tracking speed of 40.3 frames per second (FPS). On the RGBT234 dataset, the algorithm achieved a PR of 75.3% and an SR of 52.6%, with a tracking speed of 34.7 FPS. Compared with other algorithms, DuSiamIE exhibits a slight loss in accuracy but significantly outperforms them in speed on resource-constrained edge devices. It is the only algorithm among those tested that can perform real-time tracking on such devices. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results