- Acoustic Wave Phenomena Research
- Noise Effects and Management
- Layered Double Hydroxides Synthesis and Applications
- Clay minerals and soil interactions
- Multimodal Machine Learning Applications
- Underwater Acoustics Research
- Wind and Air Flow Studies
- Domain Adaptation and Few-Shot Learning
- Adsorption and biosorption for pollutant removal
- Face recognition and analysis
- Human Pose and Action Recognition
- Composite Structure Analysis and Optimization
- Advanced Battery Technologies Research
- Vibration and Dynamic Analysis
- Aerodynamics and Acoustics in Jet Flows
- Mesoporous Materials and Catalysis
- Pigment Synthesis and Properties
- Optical Coatings and Gratings
- Video Surveillance and Tracking Methods
- Vehicle Noise and Vibration Control
- Advancements in Battery Materials
- Advanced optical system design
- biodegradable polymer synthesis and properties
- Iron oxide chemistry and applications
- Minerals Flotation and Separation Techniques
Huazhong University of Science and Technology
2016-2025
China Automotive Technology and Research Center
2023-2024
University of Science and Technology Beijing
2023
Naval University of Engineering
2023
Baidu (China)
2023
Chinese Academy of Sciences
2017-2020
Lanzhou Institute of Chemical Physics
2017-2020
University of Chinese Academy of Sciences
2016-2020
Nanjing University of Science and Technology
2018
Jilin University
2016
Existing works have designed end-to-end frameworks based on Faster-RCNN for person search. Due to the large receptive fields in deep networks, feature maps of each proposal, cropped from stem maps, involve redundant context information outside bounding boxes. However, search is a fine-grained task which needs accurate appearance information. Such can make model fail focus persons, so learned representations lack capacity discriminate various identities. To address this issue, we propose...
Person detection networks have been widely used in person search. These detectors discriminate persons from the background and generate proposals of all a gallery scene images for each query. However, such large number negative influence on following identity matching process because many distractors are involved. In this paper, we propose new network search, named Instance Guided Proposal Network (IGPN), which can learn similarity between query proposals. Thus, decrease according to scores....
Deep learning based methods have achieved remarkable progress in action recognition. Existing works mainly focus on designing novel deep architectures to achieve video representations for Most treat sampled frames equally and average all the frame-level predictions at testing stage. However, within a video, discriminative actions may occur sparsely few most other are irrelevant ground truth even lead wrong prediction. As result, we think that strategy of selecting relevant would be further...
Recently large-scale language-image models (e.g., text-guided diffusion models) have considerably improved the image generation capabilities to generate photorealistic images in various domains. Based on this success, current editing methods use texts achieve intuitive and versatile modification of images. To edit a real using models, one must first invert noisy latent from which an edited is sampled with target text prompt. However, most lack following: user-friendliness additional masks or...
This study presents a foam sandwich structure reinforced with carbon fiber columns (FSS-CFC), which exhibits strong mechanical and sound insulation properties. The FSS-CFC consists of two face-sheets polyvinyl chloride (PVC) core containing multiple CFC cylinders arranged in periodic array. transmission loss (STL) measured acoustic tube experiments closely aligns the finite element simulation results, validating reliability present research. Through characteristic analyses, reveals mechanism...
A series of environment-friendly clay minerals—α-Fe2O3 iron-red hybrid pigments—were prepared by a simple one-step hydrothermal reaction process using natural nanostructured silicate minerals as starting materials. The influence structure, morphology and composition different on the color properties, stability pigments was studied comparatively systematic structure characterizations with X-ray diffraction (XRD), Fourier-transform infrared spectroscopy (FTIR), scanning electron microscopy...
To overcome the poor stability of natural lutein to environmental factors, layered double hydroxide was incorporated by a green mechanical grinding process. The influences external factors (chemical reagents, heating and light) on before after being loaded were evaluated. results confirmed that mainly adsorbed surface (LDH) via chemical interaction. Compared with pure lutein, thermal decomposition lutein/LDH improved from 100 °C 300 °C, retention ratio increased about 8.64% 21.47% 96 h light...
Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI
Summary Cell‐free seminal RNA (cfs‐ ) is mixed transcripts derived from male reproductive organs, and potential biomarker for the research diagnosis of reproductive‐related diseases. However, some clinical factors, including age, asymptomatic Ureaplasma urealyticum ( UU infection, scrotal heat stress, abstinence period, storage condition semen samples, may interfere with sperm parameters measurement biomarkers. Accordingly, this study was designed to evaluate effect above factors on cfs‐ ,...
Recently large-scale language-image models (e.g., text-guided diffusion models) have considerably improved the image generation capabilities to generate photorealistic images in various domains. Based on this success, current editing methods use texts achieve intuitive and versatile modification of images. To edit a real using models, one must first invert noisy latent from which an edited is sampled with target text prompt. However, most lack following: user-friendliness additional masks or...