Deafness is a condition that results in the loss of hearing function, hindering the reception of information such as oral communication that relies on auditory senses. Consequently, individuals with hearing impairment...
详细信息
3D human pose estimation (HPE) has improved significantly through Graph Convolutional Networks (GCNs), which effectively model body part ***, GCNs have limitations, including uniform feature transformations across nod...
详细信息
Deafness is a condition that results in the loss of hearing function, hindering the reception of information such as oral communication that relies on auditory senses. Consequently, individuals with hearing impairment...
Deafness is a condition that results in the loss of hearing function, hindering the reception of information such as oral communication that relies on auditory senses. Consequently, individuals with hearing impairment experience communication barriers and may have limited or no ability to respond. One solution is the use of sign language. In Indonesia, there are two known sign languages: Sibi and Bisindo. Both serve the same function but differ in their style of movement and expression. Bisindo is considered more flexible as it conveys meaning based on the Indonesian language. However, the universal understanding of this language solution is still limited among many people. Therefore, a program is needed to facilitate translation between deaf individuals who use sign language and their counterparts who do not communicate through sign language. CNN (Convolutional Neural Network) is a deep learning algorithm used for training visual input data recognition by computer systems. There are various CNN-based architectures, and one of them is AlexNet. Based on the author's testing, the AlexNet architecture proves to be suitable for real-time sign language translation. The evaluation of the system involved 7,800 datasets and 520 testing instances, with an average accuracy of 468 correct translations. When averaged, the system achieved a 90% accuracy rate, representing a 100% increase in accuracy compared to previous approaches.
作者：
Bimm, MorganGalloway, KateSkjerseth, AmyMorgan Bimm is an Assistant Professor of Women’s and Gender Studies at St. Francis Xavier University. Her research focuses on integrating fan studies
popular music studies and feminist theory particularly as they relate to the consumption and framing of popular culture music and aesthetics. Broadly her work is interested in the ways that technology and media work together to produce ideas about cultural relevance and gender. Morgan’s academic writing has appeared in Punk & Post-Punk Flow MAI: Feminism & Visual Culture as well as a number of scholarly anthologies. She also serves as co-chair for the Gender and Feminisms Caucus of the Society for Cinema and Media Studies (SCMS). Kate Galloway is Assistant Professor at Rensselaer Polytechnic Institute. Her in-progress monograph Remix
Reuse Recycle: Music Media Technologies and Remediating the Environment examines how and why contemporary artists remix and recycle sounds music and texts encoded with environmental knowledge. She has co-edited two special journal issues (American Music and Twentieth-Century Music) with K. E. Goldschmitt and Paula Harper that address the creative and social phenomena of internet music communities and practices of listening to the internet and edited a special issue on “Listening to/with Game Worlds” for Music and the Moving Image. With Paula Harper and Steven Gamble she co-organized the in-person/virtual Music and the Internet Conference (2023) and with Harper and Christa Bentley she is co-organizer of the Taylor Swift Study Day (2021). With Harper and Bentley she is co-editing the forthcoming collection Taylor Swift: The Star The Songs The Fans. Amy Skjerseth is Lecturer in Audiovisual Media and co-director of the Music and Audiovisual Media MA program at the University of Liverpool. Her monograph-in-progress
Audiovisual Thinking: Visual Waves of Popular Music explores how technologies including 1960s transistor radios 1990s Auto-Tune and present-day vocaloids and deepfakes have influenced both musical and visual
BigNeuron is an open community bench-testing platform with the goal of setting open standards for accurate and fast automatic neuron tracing. We gathered a diverse set of image volumes across several species that is r...
BigNeuron is an open community bench-testing platform with the goal of setting open standards for accurate and fast automatic neuron tracing. We gathered a diverse set of image volumes across several species that is representative of the data obtained in many neuroscience laboratories interested in neuron tracing. Here, we report generated gold standard manual annotations for a subset of the available imaging datasets and quantified tracing quality for 35 automatic tracing algorithms. The goal of generating such a hand-curated diverse dataset is to advance the development of tracing algorithms and enable generalizable benchmarking. Together with image quality features, we pooled the data in an interactive web application that enables users and developers to perform principal component analysis, t-distributed stochastic neighbor embedding, correlation and clustering, visualization of imaging and tracing data, and benchmarking of automatic tracing algorithms in user-defined data subsets. The image quality metrics explain most of the variance in the data, followed by neuromorphological features related to neuron size. We observed that diverse algorithms can provide complementary information to obtain accurate results and developed a method to iteratively combine methods and generate consensus reconstructions. The consensus trees obtained provide estimates of the neuron structure ground truth that typically outperform single algorithms in noisy datasets. However, specific algorithms may outperform the consensus tree strategy in specific imaging conditions. Finally, to aid users in predicting the most accurate automatic tracing results without manual annotations for comparison, we used support vector machine regression to predict reconstruction quality given an image volume and a set of automatic tracings.
暂无评论