A5101424754- Generative Adversarial Networks and Image Synthesis
- Cervical Cancer and HPV Research
- Artificial Intelligence in Healthcare and Education
- Advanced Image Processing Techniques
- Medical Imaging and Analysis
- Digital Media Forensic Detection
- Multimodal Machine Learning Applications
- Computer Graphics and Visualization Techniques
- AI in cancer detection
- Radiomics and Machine Learning in Medical Imaging
Netflix (United States)
2024
Binzhou University
2024
Binzhou Medical University
2024
Salesforce (United States)
2023
Text-to-image generation models that generate images based on prompt descriptions have attracted an increasing amount of attention during the past few months. Despite their encouraging performance, these raise concerns about misuse generated fake images. To tackle this problem, we pioneer a systematic study detection and attribution by text-to-image models. Concretely, first build machine learning classifier to detect various We then attribute source models, such model owners can be held...
Twenty-five percent of cervical cancers are classified as endocervical adenocarcinomas (EACs), which comprise a highly heterogeneous group tumors. A histopathologic risk stratification system known the Silva pattern was developed based on morphology. However, accurately classifying such patterns can be challenging. The study objective to develop deep learning pipeline (Silva3-AI) that automatically analyzes whole slide image–based images and identifies with high accuracy. Initially, total...
Masked Image Modeling (MIM) has achieved significant success in the realm of self-supervised learning (SSL) for visual recognition. The image encoder pre-trained through MIM, involving masking and subsequent reconstruction input images, attains state-of-the-art performance various downstream vision tasks. However, most existing works focus on improving MIM.In this work, we take a different angle by studying pre-training data privacy MIM. Specifically, propose first membership inference...
We present a novel framework for free-viewpoint facial performance relighting using diffusion-based image-to-image translation. Leveraging subject-specific dataset containing diverse expressions captured under various lighting conditions, including flat-lit and one-light-at-a-time (OLAT) scenarios, we train diffusion model precise control, enabling high-fidelity relit images from inputs. Our includes spatially-aligned conditioning of captures random noise, along with integrated information...