Robotics Biomim
Department of Mechanical Engineering, University of Bath, Claverton Down, Bath, BA27AY UK.
Published: April 2017
This paper explores ways of combining vision and touch for the purpose of object recognition. In particular, it focuses on scenarios when there are few tactile training samples (as these are usually costly to obtain) and when vision is artificially impaired. Whilst machine vision is a widely studied field, and machine touch has received some attention recently, the fusion of both modalities remains a relatively unexplored area. It has been suggested that, in the human brain, there exist shared multi-sensorial representations of objects. This provides robustness when one or more senses are absent or unreliable. Modern robotics systems can benefit from multi-sensorial input, in particular in contexts where one or more of the sensors perform poorly. In this paper, a recently proposed tactile recognition model was extended by integrating a simple vision system in three different ways: vector concatenation (vision feature vector and tactile feature vector), object label posterior averaging and object label posterior product. A comparison is drawn in terms of overall accuracy of recognition and in terms of how quickly (number of training samples) learning occurs. The conclusions reached are: (1) the most accurate system is "posterior product", (2) multi-modal recognition has higher accuracy to either modality alone if all visual and tactile training data are pooled together, and (3) in the case of visual impairment, multi-modal recognition "learns faster", i.e. requires fewer training samples to achieve the same accuracy as either other modality.
Download full-text PDF |
Source |
---|---|
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5395591 | PMC |
http://dx.doi.org/10.1186/s40638-017-0058-2 | DOI Listing |
Proc Natl Acad Sci U S A
January 2025
Department of Psychology, City College, City University of New York, New York, NY 10031.
Looking at the world often involves not just seeing things, but feeling things. Modern feedforward machine vision systems that learn to perceive the world in the absence of active physiology, deliberative thought, or any form of feedback that resembles human affective experience offer tools to demystify the relationship between seeing and feeling, and to assess how much of visually evoked affective experiences may be a straightforward function of representation learning over natural image statistics. In this work, we deploy a diverse sample of 180 state-of-the-art deep neural network models trained only on canonical computer vision tasks to predict human ratings of arousal, valence, and beauty for images from multiple categories (objects, faces, landscapes, art) across two datasets.
View Article and Find Full Text PDFMatern Child Health J
January 2025
School of Medicine and Public Health, University of Newcastle, Newcastle, NSW, Australia.
Objectives: Women with preconception anxiety and/or depression experience high rates of relapse or recurrence of the disorders in the perinatal period. This review aimed to identify perinatal interventions that were designed to prevent relapse or recurrence in women with a history of anxiety and/or depression.
Methods: The review was conducted based on the PRISMA guidelines.
Matern Child Health J
January 2025
Tanzania Field Epidemiology and Laboratory Training Program, Tanzania Ministry of Health, Dodoma, Tanzania.
Introduction: Population risk for neural tube defects (NTDs) can be determined using red blood cell (RBC) folate. However, a paucity of biomarker and surveillance data among non-lactating, non-pregnant women of reproductive age (NPWRA) from Africa limits accurate assessment. Our study assessed folate and vitamin B12 status among non-lactating NPWRA and predicted population risk of NTDs in Tanzania.
View Article and Find Full Text PDFCogn Res Princ Implic
January 2025
Department of Psychology and Centre for Integrative and Applied Neuroscience, York University, 4700 Keele St., Toronto, ON, M3J 1P3, Canada.
Developing ways to predict and encourage vaccine booster uptake are necessary for durable immunity responses. In a multi-nation sample, recruited in June-August 2021, we assessed delay discounting (one's tendency to choose smaller immediate rewards over larger future rewards), COVID-19 vaccination status, demographics, and distress level. Participants who reported being vaccinated were invited back one year later (n = 2547) to report their willingness to receive a booster dose, along with reasons for their decision.
View Article and Find Full Text PDFClin Oral Investig
January 2025
Institute of Science and Technology, Division of Periodontics, São Paulo State University (Unesp), Av. Eng. Francisco José Longo, 777, São José dos Campos, São Paulo, 12245-000, Brazil.
Objective: This study aimed to compare the salivary protein profile in individuals with Type 2 Diabetes Mellitus (DM2) and periodontitis and their respective controls.
Methods: Eighty participants were included in the study. The four groups were formed by individuals with DM2 and periodontitis (DM2 + P, n = 20), DM2 without periodontitis (DM2, n = 20), periodontitis without DM2 (P, n = 20) and individuals without periodontitis and without DM2 (H, n = 20).
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!
© LitMetric 2025. All rights reserved.