HazeCLIP uses language to improve dehazing methods for real-world images.
― 5 min read
Computer Science - Computer Vision and Pattern Recognition
RSSA new model improves machine recognition of unseen object-attribute combinations.
― 5 min read
This study enhances methods for detecting out-of-distribution examples in medical imaging.
― 4 min read
Introducing a method to enhance AI system resilience through multi-task adversarial attacks.
― 5 min read
MeshSegmenter enhances 3D model segmentation using textures and innovative methods.
― 7 min read
A new model for understanding 3D environments using text-based descriptions.
― 4 min read
A new method creates high-quality images from layouts using no extensive datasets.
― 6 min read
A new model combining Unet and TransUnet for improved nuclei segmentation.
― 5 min read
This article tackles miscalibration issues in vision-language models and offers solutions.
― 5 min read
A novel approach to enhance detail and quality in 3D models from text.
― 6 min read
Qalam offers improved recognition for Arabic text and handwriting.
― 6 min read
Dynamic Semantic Adjuster improves self-supervised learning performance across various tasks.
― 5 min read
Introducing new algorithms for robust plane adjustment in 3D applications.
― 8 min read
GPSFormer significantly improves understanding of 3D shapes in various applications.
― 5 min read
Combatting misleading information through new methods and technologies.
― 4 min read
New methods enhance action recognition in visual data with skeleton analysis.
― 4 min read
Study assesses nnU-Net's effectiveness in segmenting cardiac MRI images.
― 7 min read
A new benchmark sheds light on hallucination in vision language models.
― 5 min read
CycleMix enhances AI models by mixing image styles for better performance.
― 6 min read
A framework improves the conversion of sketches into CAD files, enhancing design efficiency.
― 5 min read
A new module improves robot navigation by estimating uncertainty in image segmentation.
― 6 min read
This article explores how robots perceive and interact with their environment.
― 6 min read
A novel approach to predict how people visually search for objects.
― 6 min read
DACCA enhances lane detection through improved feature learning and context aggregation.
― 7 min read
Using technology to improve emergency medical procedures and support responders.
― 6 min read
Unified-EGformer improves image quality under varying lighting conditions.
― 5 min read
A new method enhances visual odometry for underwater vehicles.
― 6 min read
A study develops a model to better identify faint galactic features in images.
― 6 min read
A new method enhances drone inspections by optimizing viewpoint selection.
― 5 min read
Examining the rise of few-shot action recognition in video analysis.
― 8 min read
A new method improves early detection of coffee leaf rust using low-quality images.
― 5 min read
FedDM enhances federated learning for diffusion models while ensuring data privacy.
― 5 min read
MetaAug reduces overfitting in PTQ through innovative data transformations.
― 6 min read
CrowdMAC improves predictions in crowd density forecasting despite incomplete data.
― 6 min read
A model designed for creating large, high-quality images efficiently.
― 6 min read
A new model improves clinical tasks in digital pathology.
― 6 min read
A new technique enhances scene classification using hybrid graph neural networks.
― 6 min read
This article discusses a new dataset for enhancing safety in human-robot teamwork.
― 7 min read
Introducing ESCAPE, a framework enhancing 3D human pose accuracy and speed.
― 6 min read
This study evaluates CNN and Modified VGG16 models on emotion recognition tasks.
― 7 min read