This presentation summarizes a paper that proposes improving supervised regression by reformulating it as a classification problem inspired by clustering approaches, and introduces a range of novel methods, including a unified end-to-end framework.
This presentation summarises the key takeaways from the paper MapAnything: Universal Feed-Forward Metric 3D Reconstruction.
This presentation summarizes the paper “Energy-Based Transformers are Scalable Learners and Thinkers”, which combines energy-based models with standard Transformers to let neural networks iteratively refine and self-verify their predictions.
The presentation will cover the motivation and goals of the algorithm, the implementation and mathematical details, example results, potential applications, and future work.
This talk provides a summary of the works presented at CVPR2024 and some spotlight papers we identified.
We will present a deep learning-based method that jointly detects and pairs nuclei and Golgi in 3D that outperforms classical approaches on a dataset of mouse retinal microscopy images acquired at the Vascular Biology and Disease Laboratory from Instituto Gulbenkian de Ciência.
Emergency Departments (EDs) are essential for providing prompt care but often face overcrowding issues, mainly due to non-urgent patients and
Automated Analysis of Multi-Modal Medical Data using Deep Belief Networks Recently, magnetic resonance and ultrasound imaging have found utility as
Precision Radiology In this work, we propose new prognostic methods that predict 5-year mortality in elderly individuals using chest computed
Started in 2014 we deployed a large scale 3D visual sensing infrastructure in Lisbon’s Airport. Currently with more than 30
