AVI-Talking creates lifelike 3D faces that express emotions through audio.
― 6 min read
Computer Science - Computer Vision and Pattern Recognition
RSSNew techniques enhance quantization while managing outliers for better model performance.
― 5 min read
A new framework enhances robot actions through human commands.
― 6 min read
Research explores enhancements in adapting models without source data access.
― 7 min read
New methods reduce human labeling while improving object detection accuracy.
― 7 min read
Simplifying lane detection through innovative sequence generation.
― 6 min read
Exploring the potential of video generation in real-world tasks.
― 6 min read
Combining RGB and Depth data improves action recognition in robotic systems.
― 6 min read
A new method automates 3D character modeling from concept art.
― 7 min read
New method enhances thermal infrared tracking performance through motion integration.
― 8 min read
A new method enhances machine learning of audio-visual data.
― 5 min read
Introducing a new method for learning object behavior in videos and 3D scenes.
― 6 min read
A look at how DIC measures asphalt concrete performance under stress.
― 7 min read
A new dataset and recovery framework aim to improve corrupted video restoration methods.
― 5 min read
New methods simplify creating distinct 3D scenes from text descriptions.
― 5 min read
ConSept framework enhances semantic segmentation by reducing forgetting in models.
― 6 min read
This study presents a new approach for segmenting the sigmoid colon in CT images.
― 6 min read
New methods improve retinal blood flow measurements for better eye disease diagnosis.
― 6 min read
A new technique improves identification of aircraft in low-quality images.
― 5 min read
Exploring advances in Zero-Shot Hashing for effective image searches.
― 7 min read
An overview of diffusion models and their impact on generative AI.
― 7 min read
BLO-SAM improves semantic segmentation with bi-level optimization and reduced manual input.
― 7 min read
Examining limitations of large vision-language models in detailed image understanding.
― 6 min read
A method to enhance GAN performance using unbalanced data.
― 7 min read
LEXIS helps robots recognize indoor spaces using language and map data.
― 4 min read
Exploring the latest methods in human shape and clothing technology.
― 8 min read
OpenMEDLab enhances access to medical AI tools and resources for better healthcare.
― 6 min read
A new method enhances segmentation accuracy using class activation maps.
― 5 min read
CAD-SIGNet improves how we reconstruct design history from point clouds.
― 5 min read
A novel method for generating images from sketches presents new creative possibilities.
― 6 min read
Machine learning techniques enhance the detection of defects in solar cells using electroluminescence images.
― 6 min read
New dataset enhances computer agents' ability to perform various tasks.
― 6 min read
Examining how models learn from multiple captions and the shortcuts they find.
― 7 min read
New techniques improve deep learning model creation and security.
― 7 min read
A new method improves the alignment of LiDAR and camera data for better 3D models.
― 6 min read
A fresh method to improve text-to-image models with efficiency and quality.
― 6 min read
A new method enhances clarity in ore image segmentation for better processing.
― 6 min read
Introducing ICP-Flow for efficient scene flow estimation in autonomous vehicles.
― 9 min read
OSASIS revolutionizes image stylization while preserving original details and structure.
― 5 min read
The Re-embedded Regional Transformer enhances cancer diagnosis through innovative feature re-embedding techniques.
― 6 min read