MathWriting boosts research in recognizing handwritten mathematical expressions with diverse samples.
― 5 min read
Computer Science - Computer Vision and Pattern Recognition
RSSA fresh approach improves comparison of persistence diagrams for data analysis.
― 6 min read
A method for creating human movement in 3D environments based on text.
― 7 min read
Recent advancements enable computers to generate lifelike talking faces for various applications.
― 4 min read
A new method for creating 3D objects using shape programs and differentiable rendering.
― 6 min read
A new method enhances 3D model accuracy using multiple viewpoints.
― 5 min read
A method to adapt machine learning models without needing source data.
― 6 min read
A new framework improves efficiency in video object detection for surveillance systems.
― 9 min read
Enhancing face images while keeping the person's identity intact.
― 8 min read
Diverse features enhance models' ability to identify new data categories effectively.
― 7 min read
New model improves machine learning by separating complex data into simpler parts.
― 7 min read
This research presents a method for quicker, high-quality 3D scene generation.
― 5 min read
Vision Transformers enhance image analysis in scientific research.
― 6 min read
Revolutionary techniques for efficient image processing using optimal transport.
― 6 min read
Examining how quantization can improve neural network performance and generalization.
― 6 min read
A new method improves few-shot segmentation using visual prompts and attention mechanisms.
― 6 min read
New methods improve robustness in data recovery for computer vision applications.
― 7 min read
NCL improves interpretability and performance in machine learning tasks.
― 6 min read
MoA innovates personalized image creation with a dual-pathway system for unique outputs.
― 8 min read
New methods improve predictions of future video frames using past data and depth.
― 7 min read
Exploring the balance of real and generated data for better machine learning performance.
― 7 min read
A new model enhances activity recognition by focusing on temporal relations among actions.
― 5 min read
SPAM offers a smarter way to label video data efficiently.
― 5 min read
TROT improves accuracy in recognizing human activities by leveraging temporal relationships.
― 7 min read
SLAIM enhances real-time tracking and mapping using advanced neural networks and depth information.
― 6 min read
New methods enhance image quality in satellite imagery.
― 4 min read
A new dataset and framework for tracking soccer players using single-camera video footage.
― 8 min read
GenLFI revolutionizes live-cell imaging with unmatched speed and field of view.
― 6 min read
A look into how robots navigate environments using various technologies.
― 5 min read
A new method enhances the quality of low-resolution satellite images.
― 6 min read
A new method enhances multimodal models using shared visual prompts.
― 8 min read
Learning dense correspondences between shapes without perfect alignment.
― 5 min read
New methods improve accuracy in counting Aedes aegypti mosquito eggs.
― 7 min read
A new method streamlines image editing by focusing on noise and timing.
― 6 min read
New methods aim to improve communication for the deaf community.
― 5 min read
A guide to rotation representations and their impact on machine learning.
― 7 min read
This article presents a method for improving uncertainty estimates in TinyML applications.
― 7 min read
Edit videos easily using images as guides for changes.
― 5 min read
A novel approach to improve compressed dark images while reducing artifacts.
― 6 min read
A method that simplifies and speeds up texture editing for images.
― 5 min read