Accelerating Deep Action Recognition Networks for Real-Time Applications

In this work, the authors propose several techniques for accelerating a modern action recognition pipeline. This article reviewed several recent and popular action recognition works and selected two of them as part of the tools used for improving the aforementioned acceleration. Specifically, temporal segment networks (TSN), a convolutional neural network (CNN) framework that makes use of a small number of video frames for obtaining robust predictions which have allowed to win the first place in the 2016 ActivityNet challenge, and MotionNet, a convolutional-transposed CNN that is capable of inferring optical flow RGB frames. Together with the last proposal, this article integrated a new software for decoding videos that takes advantage of NVIDIA GPUs. This article shows a proof of concept for this approach by training the RGB stream of the TSN network in videos loaded with NVIDIA Video Loader (NVVL) of a subset of daily actions from the University of Central Florida 101 dataset.

Download Full-text

Artificial Intelligence and Convolutional Neural Network for Recognition of Human Interaction by Video from Drone

10.20944/preprints201908.0289.v1 ◽

2019 ◽

Author(s):

Ghazal Shamsipour ◽

Saied Pirasteh

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Poor Quality ◽

Motion Blur ◽

Human Interaction ◽

Support Vector ◽

Central Florida ◽

The Difference ◽

High Level ◽

The University

Recognition of the human interaction on the unconstrained videos taken from cameras and remote sensing platforms like a drone is a challenging problem. This study presents a method to resolve issues of motion blur, poor quality of videos, occlusions, the difference in body structure or size, and high computation or memory requirement. This study contributes to the improvement of recognition of human interaction during disasters such as an earthquake and flood utilizing drone videos for rescue and emergency management. We used Support Vector Machine (SVM) to classify the high-level and stationary features obtained from Convolutional Neural Network (CNN) in key-frames from videos. We extracted conceptual features by employing CNN to recognize objects from first and last images from a video. The proposed method demonstrated the context of a scene, which is significant in determining the behaviour of human in the videos. In this method, we do not require person detection, tracking, and many instances of images. The proposed method was tested for the University of Central Florida (UCF Sports Action), Olympic Sports videos. These videos were taken from the ground platform. Besides, camera drone video was captured from Southwest Jiaotong University (SWJTU) Sports Centre and incorporated to test the developed method in this study. This study accomplished an acceptable performance with an accuracy of 90.42%, which has indicated improvement of more than 4.92% as compared to the existing methods.

Download Full-text

Two-Stream Convolutional Neural Network for Skeleton-Based Action Recognition

2020 IEEE 6th International Conference on Computer and Communications (ICCC) ◽

10.1109/iccc51575.2020.9344897 ◽

2020 ◽

Author(s):

Shi Yao ◽

Wu Muqing ◽

Xu Weiyao

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Action Recognition

Download Full-text

ACRL candidates for 2018: A look at who’s running

College & Research Libraries News ◽

10.5860/crln.79.1.25 ◽

2018 ◽

Vol 79 (1) ◽

pp. 25

Author(s):

Association Of College & Research Libraries

Keyword(s):

Social Sciences ◽

State University ◽

University Of Central Florida ◽

Research Education ◽

Central Florida ◽

Reference Services ◽

Louisiana State University ◽

Associate Dean ◽

E Learning ◽

The University

Penny Beile is associate director for research, education, and engagement at the University of Central Florida, a position she has held since 2013. Prior to this, Beile served as head of the Curriculum Materials Center and interim head of reference services at the University of Central Florida (1998–2013), as head of education resources and social sciences reference librarian at Louisiana State University (1994–98), and as social sciences reference librarian at Miami (OH) University (1992–94).Karen Munro is associate dean of libraries, learning, and research services at Simon Fraser University, a position she has held since 2017. Prior to this, Munro served as head of the Portland Library and Learning Commons at the University of Oregon (2008–17), as e-learning librarian at the University of California-Berkeley (2005–08), and as literature librarian at the University of Oregon (2002–05).

Download Full-text

Emotion Recognition in Horses with Convolutional Neural Networks

Future Internet ◽

10.3390/fi13100250 ◽

2021 ◽

Vol 13 (10) ◽

pp. 250

Author(s):

Luis A. Corujo ◽

Emily Kieson ◽

Timo Schloesser ◽

Peter A. Gloor

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Intelligent Systems ◽

High Accuracy ◽

Eye Position ◽

Proof Of Concept ◽

Concept System ◽

Validation Set ◽

Fast Region ◽

Head Neck

Creating intelligent systems capable of recognizing emotions is a difficult task, especially when looking at emotions in animals. This paper describes the process of designing a “proof of concept” system to recognize emotions in horses. This system is formed by two elements, a detector and a model. The detector is a fast region-based convolutional neural network that detects horses in an image. The model is a convolutional neural network that predicts the emotions of those horses. These two elements were trained with multiple images of horses until they achieved high accuracy in their tasks. In total, 400 images of horses were collected and labeled to train both the detector and the model while 40 were used to test the system. Once the two components were validated, they were combined into a testable system that would detect equine emotions based on established behavioral ethograms indicating emotional affect through the head, neck, ear, muzzle, and eye position. The system showed an accuracy of 80% on the validation set and 65% on the test set, demonstrating that it is possible to predict emotions in animals using autonomous intelligent systems. Such a system has multiple applications including further studies in the growing field of animal emotions as well as in the veterinary field to determine the physical welfare of horses or other livestock.

Download Full-text

In-House Deep Environmental Sentience for Smart Homecare Solutions toward Ageing Society

10.36227/techrxiv.12846563 ◽

2020 ◽

Author(s):

PHILIP EASOM ◽

Ahmed Bouridane ◽

Feiyu Qiang; ◽

Li Zhang ◽

Carolyn Downs ◽

...

Keyword(s):

Neural Network ◽

Deep Learning ◽

Home Care ◽

Convolutional Neural Network ◽

Elderly People ◽

Maximum Speed ◽

Base System ◽

Ageing Society ◽

Care Workers ◽

Real Time Applications

<p>With an increasing amount of elderly people needing home care around the clock, care workers are not able to keep up with the demand of providing maximum support to those who require it. As medical costs of home care increase the quality is care suffering as a result of staff shortages, a solution is desperately needed to make the valuable care time of these workers more efficient. This paper proposes a system that is able to make use of the deep learning resources currently available to produce a base system that could provide a solution to many of the problems that care homes and staff face today. Transfer learning was conducted on a deep convolutional neural network to recognize common household objects was proposed. This system showed promising results with an accuracy, sensitivity and specificity of 90.6%, 0.90977 and 0.99668 respectively. Real-time applications were also considered, with the system achieving a maximum speed of 19.6 FPS on an MSI GTX 1060 GPU with 4GB of VRAM allocated.<i></i></p>

Download Full-text