S-SAM simplifies medical image segmentation with less expert input and improved efficiency.
― 5 min read
Computer Science - Computer Vision and Pattern Recognition
RSSNew approach improves action classification using historical context in videos.
― 6 min read
An overview of SAM and SAM2, models transforming image segmentation efficiency.
― 6 min read
Long-form answers improve accessibility for blind and low vision individuals.
― 6 min read
CryoBench helps enhance cryo-EM method evaluation with diverse datasets.
― 5 min read
A new method merges atomic force microscopy with deep learning to improve protein shape prediction.
― 5 min read
FruitNeRF uses 3D imaging technology for accurate fruit counting in agriculture.
― 5 min read
A new framework improves how models recognize actions in videos.
― 5 min read
A new model improves speed and quality in 3D shape creation.
― 5 min read
A-BDD enhances ML training for self-driving cars in poor weather.
― 6 min read
ARPA enhances Visual Word Sense Disambiguation by integrating language and image processing techniques.
― 6 min read
New model improves accuracy in skin lesion segmentation for early cancer detection.
― 5 min read
This article presents a method to incorporate uncertainty in volume visualization for better scientific data analysis.
― 7 min read
A new model for faster and effective low-light image enhancement.
― 5 min read
DEEPTalk enhances facial animations by capturing emotional expressions from speech.
― 5 min read
HOIGen introduces a new method for recognizing unseen human-object interactions.
― 6 min read
A new approach enhances crowd counting accuracy under adverse weather conditions.
― 7 min read
This framework improves action localization in videos using probabilistic representation and context.
― 5 min read
A new algorithm improves motion tracking using Dynamic Vision Sensors by mimicking eye functions.
― 5 min read
SpbTrack improves LiDAR tracking for enhanced safety and privacy.
― 5 min read
New model improves accuracy and detail in generating 3D shapes from text.
― 6 min read
A new method reduces computation time in diffusion models while maintaining output quality.
― 6 min read
A fresh take on shape reconstruction using the many-worlds perspective.
― 5 min read
New methods enhance face recognition while addressing privacy concerns.
― 6 min read
A look at how emotions influence group behavior and interactions.
― 5 min read
Improved depth estimation from endoscopic images enhances surgical precision.
― 6 min read
Study enhances robot handling of wrinkled fabrics through human behavior analysis.
― 7 min read
Introducing a method to generate unique fonts using letters and style keywords.
― 7 min read
This work focuses on efficient sign language recognition using low-cost technology.
― 8 min read
Utilizing unlabeled data to improve machine learning models through pseudo-labeling.
― 6 min read
A new method enhances lane detection accuracy for self-driving cars.
― 5 min read
A new model predicts object responses based on human interactions.
― 6 min read
This study improves computer recognition of fingerspelling in ASL for better translations.
― 8 min read
U-KAN improves crop field segmentation using advanced deep learning techniques.
― 5 min read
A new method enhances efficiency in detecting human-object interactions in images.
― 6 min read
Discover the latest developments in text-to-image models and their impact.
― 7 min read
New methods improve scene representation using multi-spectral data.
― 5 min read
New model significantly improves accuracy in cervical cancer image classification.
― 5 min read
A new model enhances 3D perception by combining depth and semantic information.
― 7 min read
This article presents a contactless method using an event camera for accurate measurements.
― 4 min read