Generic selectors
Exact matches only
Search in title
Search in content
Search in posts
Search in pages

Advanced Computer Vision: Applications & Challenges

The application of Edge computing and Edge AI are fundamental to advanced computer vision.
  • Innotech
  • |
  • December 10, 2021

Computer Vision uses Machine Learning and Deep Learning to observe and analyse its surroundings. Computer Vision has a huge impact on all industries, from transportation to healthcare. It includes techniques used in image processing, an overarching technology that enables analysis of images and videos. Advanced Computer Vision, a more complex form of Computer Vision, is an interdisciplinary field of science and technology that deals with how computers can gain high-level understanding from digital images or videos. It aims to understand and automate tasks that the human visual system can perform. The application of Edge computing and Edge AI are basically fundamental to Advanced computer vision as the Cloud is not an adequate environment for deploying AI when you require real-time solutions or fast responses. As such, deploying AI solutions to edge computing devices (Edge AI) is the only way to overcome latency limitations of centralized cloud computing. 

Applications

Healthcare
Cancer & Tumor DetectionImage detection allows scientists to pick out slight differences between cancerous and non-cancerous images, and diagnose data from MRI scans and inputted photos as malignant or benign. Detection of brain tumors in MRI scans using Deep Neural Networks. Tumor detection software can detect tumors with high accuracy and assist doctors in making diagnoses.
Rehabilitation/ PhysiotherapyThe potential for physiotherapy and rehabilitation solutions have opened up through the analysis of human motion with capabilities like pose estimation, activity and gesture recognition, and human-object interaction. Pose estimation, particularly, has been used to help analyze patient movement. Through use of such technologies cost effective and non-invasive at-home monitoring has become possible.
Medical Skill TrainingSimulation-based surgical training platforms have been developed for surgical education. Action Quality Assessment makes it possible to develop computational approaches that automatically evaluate the surgical students’ performance which yields meaningful feedback that can help individuals to improve their skills.
Transportation
Violation DetectionsCamera-based roadway monitoring systems to reduce unsafe driving behavior and automating the detection of violations such as speeding, running red lights or stop signs, wrong-way driving and making illegal turns.
Traffic Flow AnalysisWith the rise of computer vision and AI, video analytics can now be applied to traffic cameras, which can have a vast impact on Intelligent Transportation Systems and Smart Cities.
Conversion of 2D Images Into 3D ModelsThis is applied in robotics, self-driving cars, medical diagnosis and surgical operations.
Road Condition Monitoring Computer based defect detection (automated pavement distress detection using deep learning) and condition assessment are developed to monitor concrete and asphalt civil infrastructure to make more cost-effective and consistent decisions regarding the management of pavement networks.

Challenges

  • Hardware limitations (cameras for visual input, computing hardware for AI inference) 
  • The AI component of Advanced Computer Vision requires considerable computing power
  • Complexity of scaling Computer Vision Systems 
  • High level of development risk considering development time, required domain experts, and difficulties in developing a scalable infrastructure
English (UK)