Classical algorithms typically contain domain-specific insights. This makes them often more robust, interpretable, and efficient. On the other hand, deep-learning models must learn domain-specific insight from scratch from a large amount of data using gradient-based optimization techniques. To have the best of both worlds, we should make classical visual computing algorithms differentiable to enable gradient-based optimization. Computing derivatives of classical visual computing algorithms is challenging: there can be discontinuities, and the computation pattern is often irregular compared to high-arithmetic intensity neural networks. In this article, we discuss the benefits and challenges of combining classical visual computing algorithms and modern data-driven methods, with particular emphasis to my thesis, which took one of the first steps toward addressing these challenges.

Download full-text PDF

Source
http://dx.doi.org/10.1109/MCG.2022.3149550DOI Listing

Publication Analysis

Top Keywords

visual computing
16
classical visual
12
computing algorithms
12
gradient-based optimization
8
computing
5
differentiable visual
4
computing challenges
4
challenges opportunities
4
classical
4
opportunities classical
4

Similar Publications

High-Speed Sequential DNA Computing Using a Solid-State DNA Origami Register.

ACS Cent Sci

December 2024

School of Chemistry and Chemical Engineering, New Cornerstone Science Laboratory, Frontiers Science Center for Transformative Molecules, National Center for Translational Medicine, Shanghai Jiao Tong University, Shanghai, 200240, China.

DNA computing leverages molecular reactions to achieve diverse information processing functions. Recently developed DNA origami registers, which could be integrated with DNA computing circuits, allow signal transmission between these circuits, enabling DNA circuits to perform complex tasks in a sequential manner, thereby enhancing the programming space and compatibility with various biomolecules of DNA computing. However, these registers support only single-write operations, and the signal transfer involves cumbersome and time-consuming register movements, limiting the speed of sequential computing.

View Article and Find Full Text PDF

A closed-box kernel function for numerical simulation of transient heat conduction.

Sci Rep

December 2024

College of Electrical and Information Engineering, Quzhou University, Quzhou, 324000, China.

A new kernel function, termed the closed-box kernel function, has been developed to address numerical simulation of transient heat conduction in the same medium. Firstly, this method is versatile and not limited to specific industrial scenarios or designated materials. Secondly, the method solves the spatial temperature at each time point only once, eliminating the need for multiple iterations.

View Article and Find Full Text PDF

A client-server based recognition system: Non-contact single/multiple emotional and behavioral state assessment methods.

Comput Methods Programs Biomed

December 2024

School of Communication and Information Engineering, Shanghai University, 200444, Shanghai, China. Electronic address:

Background And Objectives: In the current global health landscape, there is an increasing demand for rapid and accurate assessment of mental states. Traditional assessment methods typically rely on face-to-face interactions, which are not only time-consuming but also highly subjective. Addressing this issue, this study aims to develop a client-server-based, non-contact multimodal emotion and behavior recognition system to enhance the efficiency and accuracy of mental state assessments.

View Article and Find Full Text PDF

3D photogrammetry is a cost-effective, non-invasive imaging modality that does not require the use of ionizing radiation or sedation. Therefore, it is specifically valuable in pediatrics and is used to support the diagnosis and longitudinal study of craniofacial developmental pathologies such as craniosynostosis - the premature fusion of one or more cranial sutures resulting in local cranial growth restrictions and cranial malformations. Analysis of 3D photogrammetry requires the identification of craniofacial landmarks to segment the head surface and compute metrics to quantify anomalies.

View Article and Find Full Text PDF

Purpose: Virtual reality is used as a distraction tool during medical procedures that can cause anxiety and pain. We assessed the usefulness, engagement, value and feasibility of virtual reality to help children cope with routine clinical procedures.

Design And Methods: Quality improvement study.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!