A5039024976- AI in cancer detection
- Cutaneous Melanoma Detection and Management
- Cell Image Analysis Techniques
- Human Motion and Animation
- Advanced Vision and Imaging
- Video Analysis and Summarization
- Advanced Fluorescence Microscopy Techniques
- Digital Imaging for Blood Diseases
- Optical Coherence Tomography Applications
- Digital Media and Visual Art
- Molecular Communication and Nanonetworks
- Satellite Image Processing and Photogrammetry
- Advanced Measurement and Detection Methods
- Advanced biosensing and bioanalysis techniques
- Gait Recognition and Analysis
- Genital Health and Disease
- Surface Roughness and Optical Measurements
- Advanced Fiber Optic Sensors
- Infrared Target Detection Methodologies
- Computer Graphics and Visualization Techniques
- Human Pose and Action Recognition
- Concrete Corrosion and Durability
- Anomaly Detection Techniques and Applications
- Image Processing Techniques and Applications
Northeast Petroleum University
2024-2025
Peking University
2018-2024
East China University of Science and Technology
2018-2019
Designing a camera motion controller that has the capacity to move virtual automatically in relation with contents of 3D animation, cinematographic and principled way, is complex challenging task. Many rules exist, yet practice shows there are significant stylistic variations how these can be applied. In this paper, we propose an example-driven which extract behaviors from example film clip re-apply extracted through learning collection motions. Our first technical contribution design...
Melanoma is the most deadly form of skin cancer worldwide. Many efforts have been made for early detection melanoma with deep learning based on dermoscopic images. It crucial to identify specific lesion patterns accurate diagnosis melanoma. However, common are not consistently present and cause sparse label problems in data. In this paper, we propose a multi-task U-Net model automatically detect attributes The network includes two tasks, one classification task classify if present, other...
Abstract Designing effective camera trajectories in virtual 3D environments is a challenging task even for experienced animators. Despite an elaborate film grammar, forged through years of experience, that enables the specification motions cinematographic properties (framing, shots sizes, angles, motions), there are endless possibilities deciding how to place and move cameras with characters. Dealing these part complexity problem. While numerous techniques have been proposed literature...
Abstract Skin lesion is a severe disease globally. Early detection of melanoma in dermoscopy images significantly increases the survival rate. However, accurate recognition skin extremely challenging manually visualization. Hence, reliable automatic classification lesions meaningful to improve pathologists’ accuracy and efficiency. In this paper, we proposed two-stage method combine deep learning features clinical criteria representations address automated diagnosis task. Stage 1 - feature...
Melanoma is a type of skin cancer with the most rapidly increasing incidence. Early detection melanoma using dermoscopy images significantly increases patients' survival rate. However, accurately classifying lesions by eye, especially in early stage melanoma, extremely challenging for dermatologists. Hence, discovery reliable biomarkers will be meaningful diagnosis. In recent years, value deep learning empowered computer-assisted diagnose has been shown biomedical imaging-based decision...
Malignant melanoma is one of the leading cancers around world. It critical to timely diagnose and treat improve patient survival. This paper proposes a deep learning model C-UNet for skin lesion segmentation. The incorporates Inception-like convolutional block, recurrent block dilated layers. We also apply finetune technique using Dice loss after training with commonly used cross-entropy loss. conditional random field was further smooth predicted label maps. Experiment results show that...
We present a novel technique that enables 3D artists to synthesize camera motions in virtual environments following style , while enforcing user-designed keyframes as constraints along the sequence. To solve this constrained motion in-betweening problem, we design and train generator from collection of temporal cinematic features (camera actor motions) using conditioning on target keyframes. further condition with code control how perform interpolation between Style codes are generated by...
Melanoma is a type of skin cancer with the most rapidly increasing incidence. Early detection melanoma using dermoscopy images significantly increases patients' survival rate. However, accurately classifying lesions by eye, especially in early stage melanoma, extremely challenging for dermatologists. Hence, discovery reliable biomarkers will be meaningful diagnosis. Recent years, value deep learning empowered computer-assisted diagnose has been shown biomedical imaging based decision making....
Abstract —This paper proposes an innovative method for Part1, skin lesion segmentation of the ISIC 2018 Challenge. Our network C-UNet is based on UNet network, we combined several methods this basic which made some improvements Jaccard Index ultimately, our yield average 0.77 On-line validation dataset.
ABSTRACT Melanoma is the most deadly form of skin cancer world-wide. Many efforts have been made for early detection melanoma. The International Skin Imaging Collaboration (ISIC) hosted 2018 Challenges to improve diagnosis melanoma based on dermoscopic images. In this paper, we describe our solution task 2 ISIC Challenges. We present a multi-task U-Net model automatically detect lesion attributes Our deep learning achieves Jaccard index 0.433 official test data, which ranks 5th place final...
Designing effective camera trajectories in virtual 3D environments is a challenging task even for experienced animators. Despite an elaborate film grammar, forged through years of experience, that enables the specification motions cinematographic properties (framing, shots sizes, angles, motions), there are endless possibilities deciding how to place and move cameras with characters. Dealing these part complexity problem. While numerous techniques have been proposed literature...