Sci

8 papers8 directions
Sci is a research venue that published work in the fields of machine learning and artificial intelligence between 2019 and 2022. The venue's contributions focus on the application of computer vision and deep learning techniques to diverse domains, ranging from medical diagnostics to cultural heritage. Research published here often explores the practical implementation of neural networks and the interpretation of experimental findings in specialized scientific and technical contexts. A significant portion of the research at Sci involves the analysis of historical artifacts, particularly ancient Roman coins. Published papers have detailed methods for classification by denomination, visual reconstruction using cycle-consistent generative adversarial networks (CycleGANs), and the curation of specific datasets for computer vision evaluation in numismatics. Other research areas include the development of biologically plausible deep neural network learning and the real-time 3D conversion of Japanese Ukiyo-e art. The venue also features work on the nuances of interpreting X-ray analysis through deep learning, emphasizing the importance of reporting standards in experimental findings. Prolific contributors to the venue include Ognjen Arandjelovic, Marios Zachariou, and Neofytos Dimitriou. Through these diverse applications, Sci serves as a platform for interdisciplinary research that integrates computational methods with specific domain expertise in history, art, and medicine.

Papers by Year

2022 2
2020 5
2019 1

Top Research Directions

Research directions most represented by papers from this venue.

Multi Metric Evaluation for Image Segmentation 1
Moving Beyond Concatenation in Multi Scale Feature Fusion 1
Luminance Chrominance Decoupling for Image Restoration and Synthesis 1
Eliminating Spurious Feature Reliance in Medical Imaging Models 1
Multi-Metric Evaluation of Clinical Prediction Models 1
The Bag of Visual Words Representation 1
Learning to Translate Images Without Paired Data 1
Gradient Weighted Class Activation Mapping 1

Top Authors

AuthorPapers
Ognjen Arandjelovic3
Haoyu Wang1

Impactful Papers

Papers from this venue that shaped research clusters.

Influential

Papers

TitleYear
Nuances of Interpreting X-ray Analysis by Deep Learning and Lessons for Reporting Experimental Findings2022
Towards New Generation, Biologically Plausible Deep Neural Network Learning2022
Classification of Ancient Roman Coins by Denomination Using Colour, a Forgotten Feature in Automatic Ancient Coin Analysis2020
Images of Roman Imperial Denarii: A Curated Data Set for the Evaluation of Computer Vision Algorithms Applied to Ancient Numismatics, and an Overview of Challenges in the Field2020
Learning to Describe: A New Approach to Computer Vision Based Ancient Coin Analysis2020
Making Japenese Ukiyo-e Art 3D in Real-Time2020
Visual Reconstruction of Ancient Coins Using Cycle-Consistent Generative Adversarial Networks2020
Superpower: X-Ray Vision2019