Computer Vision-based Social Distancing Surveillance Solution with Optional Automated Camera Calibration for Large Scale Deployment

• URL: http://arxiv.org/abs/2104.10891v1
• Date: Thu, 22 Apr 2021 06:43:02 GMT
• Title: Computer Vision-based Social Distancing Surveillance Solution with Optional Automated Camera Calibration for Large Scale Deployment
• Authors: Sreetama Das (1), Anirban Nag (1), Dhruba Adhikary (1), Ramswaroop Jeevan Ram (1), Aravind BR (1), Sujit Kumar Ojha (1), Guruprasad M Hegde (2) ((1) Engineering Data Sciences, (2) Research and Technology Centre, Robert Bosch Engineering and Business Solutions Private Limited, Koramangala, Bangalore, India)
• Abstract summary: We describe a computer vision-based AI-assisted solution to aid compliance with social distancing norms. The solution consists of modules to detect and track people and to identify distance violations.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Social distancing has been suggested as one of the most effective measures to break the chain of viral transmission in the current COVID-19 pandemic. We herein describe a computer vision-based AI-assisted solution to aid compliance with social distancing norms. The solution consists of modules to detect and track people and to identify distance violations. It provides the flexibility to choose between a tool-based mode or an automated mode of camera calibration, making the latter suitable for large-scale deployments. In this paper, we discuss different metrics to assess the risk associated with social distancing violations and how we can differentiate between transient or persistent violations. Our proposed solution performs satisfactorily under different test scenarios, processes video feed at real-time speed as well as addresses data privacy regulations by blurring faces of detected people, making it ideal for deployments.

Related papers

• Log-Likelihood Score Level Fusion for Improved Cross-Sensor Smartphone Periocular Recognition [52.15994166413364]
  We employ fusion of several comparators to improve periocular performance when images from different smartphones are compared. We use a probabilistic fusion framework based on linear logistic regression, in which fused scores tend to be log-likelihood ratios. Our framework also provides an elegant and simple solution to handle signals from different devices, since same-sensor and cross-sensor score distributions are aligned and mapped to a common probabilistic domain.
  arXiv  Detail & Related papers  (2023-11-02T13:43:44Z)
• Unsupervised domain-adaptive person re-identification with multi-camera constraints [0.0]
  We propose an environment-constrained adaptive network for reducing the domain gap. The proposed method incorporates person-pair information without person identity labels obtained from the environment into the model training. We develop a method that appropriately selects a person from the pair that contributes to the performance improvement.
  arXiv  Detail & Related papers  (2022-10-25T13:12:28Z)
• Deep Learning Techniques for Visual Counting [0.13537117504260618]
  We investigated and enhanced Deep Learning (DL) techniques for counting objects in still images or video frames. In particular, we tackled the challenge related to the lack of data needed for training current DL-based solutions.
  arXiv  Detail & Related papers  (2022-06-07T06:20:40Z)
• Feeling of Presence Maximization: mmWave-Enabled Virtual Reality Meets Deep Reinforcement Learning [76.46530937296066]
  This paper investigates the problem of providing ultra-reliable and energy-efficient virtual reality (VR) experiences for wireless mobile users. To ensure reliable ultra-high-definition (UHD) video frame delivery to mobile users, a coordinated multipoint (CoMP) transmission technique and millimeter wave (mmWave) communications are exploited.
  arXiv  Detail & Related papers  (2021-06-03T08:35:10Z)
• Adaptive Autonomy in Human-on-the-Loop Vision-Based Robotics Systems [16.609594839630883]
  Computer vision approaches are widely used by autonomous robotic systems to guide their decision making. High accuracy is critical, particularly for Human-on-the-loop (HoTL) systems where humans play only a supervisory role. We propose a solution based upon adaptive autonomy levels, whereby the system detects loss of reliability of these models.
  arXiv  Detail & Related papers  (2021-03-28T05:43:10Z)
• Joint Noise-Tolerant Learning and Meta Camera Shift Adaptation for Unsupervised Person Re-Identification [60.36551512902312]
  unsupervised person re-identification (re-ID) aims to learn discriminative models with unlabeled data. One popular method is to obtain pseudo-label by clustering and use them to optimize the model. In this paper, we propose a unified framework to solve both problems.
  arXiv  Detail & Related papers  (2021-03-08T09:13:06Z)
• Monitoring and Diagnosability of Perception Systems [21.25149064251918]
  We propose a mathematical model for runtime monitoring and fault detection and identification in perception systems. We demonstrate our monitoring system, dubbed PerSyS, in realistic simulations using the LGSVL self-driving simulator and the Apollo Auto autonomy software stack.
  arXiv  Detail & Related papers  (2020-11-11T23:03:14Z)
• Efficient Empowerment Estimation for Unsupervised Stabilization [75.32013242448151]
  empowerment principle enables unsupervised stabilization of dynamical systems at upright positions. We propose an alternative solution based on a trainable representation of a dynamical system as a Gaussian channel. We show that our method has a lower sample complexity, is more stable in training, possesses the essential properties of the empowerment function, and allows estimation of empowerment from images.
  arXiv  Detail & Related papers  (2020-07-14T21:10:16Z)
• The Visual Social Distancing Problem [99.69094590087408]
  We introduce the Visual Social Distancing problem, defined as the automatic estimation of the inter-personal distance from an image. We discuss how VSD relates with previous literature in Social Signal Processing and indicate which existing Computer Vision methods can be used to manage such problem.
  arXiv  Detail & Related papers  (2020-05-11T00:04:34Z)
• Unsupervised Domain Adaptation in Person re-ID via k-Reciprocal Clustering and Large-Scale Heterogeneous Environment Synthesis [76.46004354572956]
  We introduce an unsupervised domain adaptation approach for person re-identification. Experimental results show that the proposed ktCUDA and SHRED approach achieves an average improvement of +5.7 mAP in re-identification performance.
  arXiv  Detail & Related papers  (2020-01-14T17:43:52Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.